To begin with, there are as many as nine different floor classifications. The first and second are the classifications we are primarily interested in:
CLASS 1: light foot, residential surfaces mainly with coverings, graded for drainage, level and suitable for applied coverings; curing with a single trowel finish.
CLASS 2: foot traffic, offices and churches, usually with floor covering, subsurface tolerance, non-slip in specified areas, single troweling, non-slip finish where required.
Now to the important part. Based on curves that appear in most concrete text books about half of the total amount of shrinkage that will ever occur in concrete occurs in the first 60 days after casting. The remaining half occurs over 20 years plus. Virtually all shrinkage is complete within 3 years after casting.
If a floor is placed within the first several months after casting and shrinkage cracks developed before the floor is placed, there is a possibility that widening of the crack will occur after the floor is placed and the crack may telegraph through the tile if a membrane is not used.
We have to absorb the fact that concrete is a reactive material. It contracts, expands, curls, deflects, shrinks and is continually going through a chemical hardening process. It is permeable and subject to moisture intrusion from a variety of sources including below and lateral from poor landscaping.
Our primary concerns are shrinkage, cracks and permeability. Drying shrinkage begins after the concrete has become stiff and firm. This shrinkage is due to the loss of excess water needed to make the mix workable. The water needed for hydration (makes the concrete hard) is retained while the excess water is being "lost" either due to evaporation, into the forms, or below in the soil or sand blanket beneath. The water loss in effect means volume lost. This is drying shrinkage; in many cases steel reinforcing or movement joints can reduce the problem. Shrinking causes cracks.
Even though cracking is normal due to loss of mix water other factors are involved. Of course the water-cement ratio is first questioned, but the cement used, the aggregate, even the humidity and the temperature are also questioned. Cracks may also occur due to poorly compacted subgrade material. A "curled" slab may also be a form of shrinkage as it is a result of differential drying. After placement the excess water, not consumed by hydration, evaporates more rapidly from the top of the slab. The bottom remaining wet either due to the soil or the vapor retarder, (with water "ponding" on the surface). This causes the top to shrink more than the bottom of the slab and results in "curling", forming a concave shape to the slab. I would be remiss for not mentioning a new proposed standard for the Installation of Thick Poured Gypsum, concrete underlayments and preparation of the surface to receive resilient flooring. This proposed standard is currently under ballot with the ASTM under the designation WK4001.
Basically the practice covers the installation of poured gypsum concrete underlayments on a wood structural panel subfloor or over concrete floors in commercial structures. The proposed standard excludes use over concrete slabs on grade due to potential moisture problems. It is a well-written proposal and if passed everyone should seek a copy.
There is another ASTM standard namely E-710. F-710 is a standard practice for preparing concrete and other monolithic floors to receive resilient flooring. The instructions are very helpful and will aid everyone in proper prep and assist in avoiding problems.
Use of a vapor retarder is considered to be a problem and often times it will be deliberately pierced to prevent curling. I have now heard that the American Concrete Institute may withdraw their recommendation for the use of a vapor retarder due to perceived problems. This will definitely cause our industry more moisture problems. We will have to rely on a capillary beak from the gravel placed beneath the slab.
Concrete "bleeding" is another problem for our industry. This occurs during the consolidation of the "wet" concrete. The heavier aggregates settle to the bottom of the slab while what remains on top is a cement paste. Too much "bleeding" can result in a weak surface and scaling. "Bleeding" also results in forming a surface powder that contains alkaline salts from the cement, which by nature is alkaline. These alkaline salts can attack your floor covering adhesive as the water in the adhesive activates these salts. The best way to remove these dry salts by using water.
All in all there is no question that proper floor prep is required in all of the above situations. We cannot afford complaints and call backs.