9.4 Invalid Results and Retest Decisions

Treat Invalid Results as Quality Events

A field test can be performed with serious effort and still produce an invalid or unusable result. The technician may discover that the sample was not representative, a slump specimen sheared, an air meter leaked, the volumetric air reading did not stabilize as required, the density measure was not properly filled, or strength specimens were damaged during initial curing. The right response is not to force the number into the report as if nothing happened.

Invalid results matter because project decisions can be made from field data. A false low slump, false air reading, or mislabeled specimen set can trigger arguments, wasted concrete, missed corrective action, or misleading strength interpretation. Recognizing invalidity is therefore a professional skill, not a side issue.

The first step is to identify the kind of problem. If the issue is procedural and the method allows a repeat using another portion of the same controlled sample, do that promptly. If the sample itself is compromised, a repeat from that sample may not solve the problem. If the equipment is not functioning, continuing to generate numbers only creates a larger record problem.

Timing problems need special attention. Fresh concrete changes. If the technician discovers a delay after the sample sat unprotected, the correct answer is not to hurry through the remaining tests and pretend they were timely. Document the delay, notify the required person, and follow project and method requirements for obtaining a valid sample. A delayed invalid result is still invalid even if it is written neatly.

Invalid-test communication should be factual. State what happened, what method step or quality condition was affected, whether a repeat test was performed, and who was notified. Avoid exaggeration. Do not write that a load failed because a cone tipped over before the test was completed. Do write that the slump test was invalid due to disturbance of the cone and that a repeat test was performed on a representative portion when available.

Use this decision list when a result seems questionable:

1. Stop and identify whether the issue is sample, equipment, procedure, timing, calculation, or documentation.
2. Check the applicable method or project procedure for repeat-test requirements.
3. Do not average a valid result with an invalid result.
4. Do not adjust a reading to match the expected value.
5. Record the invalid condition and any repeat test clearly.
6. Notify the required QA/QC contact before the information is needed for acceptance decisions.
7. Remove, repair, or replace equipment that cannot produce valid results.

The written exam may describe a tempting shortcut, such as reporting the closest expected number after a meter leak or ignoring a sheared slump because the second result looks better. Those answers should raise concern. Quality control depends on using valid results and showing why any questionable result was not used.