What should I do when PVA yarn lab results disagree?

Pause before treating two numbers as a contradiction

Two reports can disagree because the yarn differs, because the specimens differ, or because the laboratories did not run the same procedure. Before declaring a failed lot, compare the sample source, package identity, conditioning, test preparation, machine setup, units, calculation, and reported endpoint. A method name by itself is not enough if the parties selected different options within it.

For a linear-density comparison, write down how each lab prepared, conditioned, and weighed its specimen, along with the unit and calculation basis. If those choices differ, the reports should not be judged as though they describe identical test conditions; agree a common procedure before ordering a comparison.

Build a fair comparison from one identified lot

Ask each lab for its full report and retain the tested specimens or remaining material. Select a fresh group of specimens from the same identified lot, make them as comparable as the method permits, and divide them between the laboratories using a documented assignment. Keep the package IDs and preparation record with each group. If the original material cannot be traced to one lot, resolve that identity gap before comparing performance.

Agree how to evaluate the comparison

When a difference could change a commercial decision, agree a comparison plan before the new tests begin. Use specimens from the same traceable lot, make them as alike as the agreed procedure permits, and record how they are assigned to each lab. If the decision depends on whether one lab consistently reports higher or lower values, involve someone qualified to plan and interpret that comparison rather than relying on a single pair of results.

Do not average the two original results to create a convenient pass value. If a systematic lab difference is found, investigate its cause and agree how future results will be interpreted. If the discrepancy comes from sample preparation or a calculation, correct the record and rerun only the affected part of the comparison.

Keep the lot decision separate from method repair

While the comparison is underway, keep the affected material on the disposition status required by the contract. Do not use a disputed result to release unrelated lots, and do not assume that a laboratory difference proves either supplier fault or buyer error. The decision should identify which lot and quantity are covered and whether production can continue with other approved material.

For dissolution differences between production batches, use the site's batch consistency guide. This answer addresses disagreement between laboratories, not the broader question of how to compare a sequence of lots.

Close the loop in writing

The closeout should include both original reports, the agreed comparison plan, specimen allocation, raw or repeat results, method deviations, and the final material decision. If the test procedure changes, assign a new revision and tell both labs which version applies to the next order. That prevents a corrected comparison from becoming a fresh mismatch on the following shipment.

If the teams cannot align on a method or specimen identity, keep the dispute open and record that limitation. A third report from another lab will not resolve the issue unless the comparison is designed to answer the same defined question.