Laboratory data often feels factual and objective. But if the lab samples do not match the design case, the engineering basis has not yet been reconciled. That matters because sizing, uncertainty assessments, performance expectations, and operating limits may all depend on which basis is actually appropriate.
The first mistake to avoid
A common but sub-optimal reaction is to ask which one is right. The better question is what each basis was intended to represent. A design case may be conservative, end-of-life, or envelope-based. A lab sample may represent only one operating point or one moment in time.
What a mismatch can mean
1. The design case is intentionally conservative
This may be perfectly acceptable if the conservatism is documented and still credible.
2. The design case is no longer representative
If recent samples are materially more onerous than the original basis, the assumptions behind the design or reporting case may no longer hold.
3. The samples are not representative of the design scenario
This is also common. A design case may cover the full envelope while a sample reflects only one limited condition.
Why this matters
A calculation can be carried out perfectly and still produce the wrong answer if the basis is wrong. The engineering risk is often not the maths. It is the unreconciled basis behind the maths.
Analytical quality is not the same as representativeness
A lab can analyse the bottle accurately and still give you a result that is unsuitable for the design case. Analytical quality answers whether the bottle was measured correctly. The real engineering question is whether the bottle represented the conditions the case was supposed to cover.
What good practice looks like
- define what the design case actually represents
- define what the lab sample actually represents
- compare parameter by parameter rather than making broad statements
- focus on consequence, not just difference
- re-check equipment or calculations where the mismatch is materially unfavourable
Changing fluid basis? Uncertainty basis hard to explain?
MeterProof helps make assumptions and inputs explicit, which makes the discussion with reviewers and verifiers much easier.