Sampling, Test Reports, Specifications, and Ethics
Scope
Sampling, reporting, specification reading, and ethics are tested together because they answer one question: can the result be trusted and defended?
Sampling
Take samples at random locations, at the depth being represented, in the quantity the test method needs. Bag and label with project, location or station, depth, date, sample type, and method. Protect moisture-sensitive samples from drying; split samples for the owner or referee lab and keep the remainder until the report is accepted.
Reports
A defensible report shows project and sample identification, test method, date tested, result with units, the laboratory maximum density and optimum moisture used for comparison, the calculated percent compaction, and any deviation from the method. Report failing results; do not average a failing test with a passing one.
Specifications
Read the order of precedence in the project documents - usually standard specifications, supplemental specifications, special provisions, then plans. When two documents conflict on test method or required density, the higher-precedence document governs; get the resolution in writing before testing.
Ethics
Sign only for work you performed or directly supervised, keep certifications and gauge calibrations current, and never alter a number or a location to make a lot pass.
Worked example
A technician is asked to retest a failing density location "somewhere better." The correct action is to report the failing result as taken, then determine whether the location was valid - for example, if rock or a disturbed surface invalidated it - and, if so, retest at a documented adjacent location. A passing retest at a new spot does not replace the original failing data unless the specification allows it.
Traps
Reporting a result without the method or the comparison density; omitting units (pcf versus kg/m³); signing for a test you did not see run; accepting a verbal instruction that contradicts the specification.