INSTITUTIONAL RECORDSMO–RD–014
REVISION 1 / PUBLIC EDITION
INSTITUTION / Review dissent
Permission to disagree
The instrument comparison survives a change of fixtures. The disagreement concerns what a successful replacement is supposed to preserve.
Dr. Sera Noll / Methods Office; Dr. Iona Vale / Material Behavior
SOURCE DATE 2026.08.12 / RELEASE DATE ABOVE
01 / A difference that moves with the specimen
E2-A and E2-B were exchanged between three calibrated fixtures under the same declared profiles. A repeatable difference in their detailed response followed the specimen. Both remained inside the six acceptance limits. The swap weakens a fixed-fixture explanation. It does not distinguish loading history from unmeasured manufacturing variation or establish a particular microscopic mechanism.
02 / Vale’s position
The original panel has undergone months of local adaptation. Its response is useful precisely because the material participates in the calculation. We can export the declared adjustment state and fabricate the nominal structure. We cannot yet show that those two operations transfer everything the specimen has acquired. A production process must account for the difference before we call the replacement equivalent.
03 / Noll’s finding
Tern has met the written acceptance conditions. The desired equivalence extends beyond them. Calling the delivered panel a failure would move our uncertainty into the contractor’s record. The institution should accept the contracted article, pay for the work, and retain its own transfer restriction. Procurement completion and scientific closure need separate signatures.
04 / The continuing disagreement
Vale considers the specimen’s loading history a likely missing part of the fabrication package. Noll considers that a testable hypothesis alongside manufacturing variation. Both support a blinded comparison of separately fabricated panels with deliberately controlled loading histories. Neither supports solving the question by redefining equivalence after seeing which specimen performs better.
05 / Responsibility accepted
The Research Directorate authorized the contracted payment and a bounded follow-on study. It did not authorize production release. Tern may recover its held capacity. The candidate program carries the delay and its cost in its own account. The two specimens remain available to the independent custodian.