ACI MFTT common mistakes
ACI MFTT · Understand the traps and how to avoid them.
Pouring water into a mortar sample that has stiffened on the board so it can be rodded into cube molds, then reporting the result as the batch's mortar.
Specimens have to be molded quickly; a dry, crumbly mortar will not consolidate, and adding water is the fastest way to make it behave.
What to do instead
Sample at the point of use from a fresh batch, taking increments across the batch, and never retemper or add water to a test sample. If the sample stiffens before molding is complete, discard it and take a new sample from the next batch, noting on the record that the specimens came from a fresh sample.
Selecting units for compressive strength and absorption testing off the top row of the storage cube because they are easy to reach, skipping chipped or dusty ones, and not recording which lot or delivery they came from.
The top row is the easiest to reach and looks the best; nobody wants to climb the cube or pull units from the middle of a strap.
What to do instead
Select units at random from the lot by a documented method (for example, random numbers mapped over the cube), include the range of condition actually present in the lot, and record lot identification, delivery date, and where in the cube each unit came from. Never pre-select only undamaged units.
Running the cone-penetration consistency test on whatever is left in the pan after the cubes are molded, and filling in sample time and test time from memory at the end of the day.
Specimens are the visible deliverable and get made first; the consistency test is treated as the leftover task done with leftover material.
What to do instead
Split the sample when it arrives. Run consistency, and take the water-content portion, from a companion portion taken at the same time as the cube sample, within the time limits of the method. Record the time the sample was taken, the time of each test, mortar temperature, and air temperature on the data sheet as the work is done.
Molding 2-in. mortar cubes in a single lift, tamping hard to close voids, stripping the molds early, and letting the cubes cure uncovered on a shelf or in a vehicle, then reporting a 7-day result as if it were the specified-age strength.
Two layers with the specified tamping take time; over-tamping one lift produces a smoother-looking cube, and the shelf is simply where the cubes are visible.
What to do instead
Use nonabsorptive steel cube molds, fill in the specified number of layers with the specified tamping pattern, strike off, cover, and cure moist in the molds; strip at the specified time, then cure in a moist environment until the test age. Label each cube with its test age and never substitute one age's result for another.
Building prisms from whatever units and mortar are on hand at the time, such as units soaked wet, mortar of a different type or from a different batch, and bed joints thinner than those in the wall, then recording only the prism strength.
Prism construction requires planning ahead; units set aside for prisms dry out, mortar type changes mid-job, and the priority becomes getting prisms built before the crew moves on.
What to do instead
Plan prism construction before the work starts. Reserve representative units conditioned as they will be in the wall, use the same mortar and grout type and the same batch source, build joints to the specified thickness, and record materials, batch, joint thickness, curing environment, and construction date on the prism record.
Testing a prism whose bearing surfaces are out of flat or not perpendicular, with a cap cast the same morning, and setting it on the lower platen without checking that it is centered.
Testing machine time is booked in a short window and caps have not had time to cure; the prism looks even enough, and the platen's center reference is hidden once the prism is on it.
What to do instead
Grind or cap to the specified flatness and perpendicularity, let caps reach adequate strength before testing, center the prism on the platen using the marked reference lines, and apply load continuously at the specified rate. Re-cap or re-grind specimens that fail the tolerances instead of testing them as they are.
Dividing the peak load by length times width of a hollow unit and reporting that value as the unit's compressive strength, without measuring the core geometry.
Length and width are on the label and take seconds; core measurements are awkward and the void-area calculation is easy to get wrong, so it gets skipped.
What to do instead
Determine net cross-sectional area from the void area measured at the critical plane for each specimen, compute strength on the basis the standard requires, and state the basis (net or gross area) explicitly alongside every result.
Weighing the absorption specimen after one oven cycle or when it merely looks dry, and wiping a soaked specimen with a towel using a different technique with each specimen, then using those masses to compute absorption.
Repeated oven cycles stretch a two-day test toward a week, and surface moisture after soaking is hard to judge the same way twice, especially when different technicians handle different specimens.
What to do instead
Return specimens to the oven until successive weighings show constant mass at the specified temperature, cool them as the method requires before weighing, and use one documented surface-drying procedure for every specimen in the set. Record every mass, not only the final one.
Computing the saturation coefficient as the 5-hour boil absorption divided by the 24-hour immersion absorption, on the reasoning that the larger number should go on top.
The two absorptions sit side by side as similar numbers in adjacent columns, and the formula is usually applied from memory at the bench.
What to do instead
Saturation coefficient equals the 24-hour immersion absorption divided by the 5-hour boil absorption. Write the formula and symbol definitions on the calculation sheet, compute only from verified dry, immersed, and boiled masses, and sanity-check that each coefficient is less than 1 before releasing the result.
Submitting a results sheet carrying only specimen strengths: no sampling location, time the sample was taken, mix or batch identification, specimen dimensions, test age, date of test, or note of any departure from the procedure.
The requester wants numbers; the field sheet is completed under time pressure, and the identification fields are deferred to later, which never arrives.
What to do instead
Use a data sheet built from the reporting section of the applicable standard for each test type, complete the identification fields at the moment of sampling, and write deviations down as they happen. A result lacking location, time, age, and dimensions cannot be used by the designer or defended if challenged.
Accepting prism testing and unit absorption work because the laboratory is short-handed, and leaving the performed-by and checked-by lines blank so the report carries no name.
Certification scope is read once at hire; the schedule says the work is due, and the technician assumes the certificate covers whatever the laboratory does.
What to do instead
Confirm which tests your certification covers before accepting an assignment, decline or escalate work outside that scope, and record the name and certificate of the person who performed the test and the person who verified it, completing the check before the report is issued.