How do fibers — steel and macrosynthetic — affect the performance of ground-supported concrete floor slabs? Everybody has an opinion, but facts are few.

Some designers base their decisions on laboratory tests that were performed on small specimens. While lab results are better than nothing, the behavior of a few cylinders or beams can only go so far in predicting the behavior of a full-sized concrete slab.

Other designers base their decisions on experience, but that approach also has its limitations. We all can tell when a slab works, but can we tell why? Was it the mix design, the reinforcement, the joint spacing, or something else? What we need are full-scale, controlled experiments, but these cost money, so there have never been many of them.

This year a rare opportunity has arisen, and a concrete contractor in North Carolina (Ace/Avant Concrete) is taking advantage of it. The contractor bought some land with five identical sheds on it. Each shed was 30 feet wide and 102 feet long (inside dimensions) with a dirt floor. The sheds will be used to store tools and materials, so the contractor decided to place a concrete slab in each. Wanting to learn more about how to control cracks and curling, the contractor decided to try something different in each shed.

Three of the slabs were cast on Oct.1, 2015, and the remaining two were cast on Oct. 30. Weather conditions were similar on both days. The concrete mix included, per cubic yard:

  • 1485 pounds #467 stone
  • 550 pounds #78 stone
  • 1279 pounds natural sand
  • 32 gallons water (somewhat more for slab 3)
  • 9 ounces Type F superplasticizer (14 ounces on slab 3)

The plan is to monitor each slab for drying shrinkage, curling, and cracking. To check for drying shrinkage, five brass surveying monuments were embedded in each slab, with lines scribed on them for length measurements. To check for curling, floor profiles were measured with a 12-inch electronic level. To check for cracking, a site-based employee inspected each slab daily for the first week, and will do so weekly after that.

A separate experiment not related to fibers is taking place on slab 4. Half this slab was finished normally and was cured with a dissipating-resin compound. The other half was treated with colloidal silica, both as a finishing aid before power floating and as a final cure. The plan is to monitor curling in both areas and to measure wear resistance after 28 days.

Early results are surprising, but interesting. As of the end of January, no cracks whatever have appeared in the four slabs that were left joint-free, including the slab without fibers. The slab with joints has cracked under three of the sawcuts. Concrete Construction will publish updates and more details as this experiment continues.

George Garber is a concrete flooring consultant with Face Consultants, Lexington, Ky.

  • Five Slabs

    Each of the five slabs in this experiment has these characteristics:

    Dimensions: 30 feet x 102 feet x 6 inches thick

  • Vapor barrier/slipsheet: 10-mil polyethylene
  • Finish: burnished with ride-on power trowels
  • Cure: dissipating-resin compound (except for half of slab 4)

  • The differences among the slabs are:

    Slab 1 included Type II steel fibers, 1-inch long, straight, at a dosage of 44.4 pcy. There were no joints.

  • Slab 2 included Type I steel fibers, 1-3/8 inches long, with hooked ends, at a dosage of 44.0 pcy. There were no joints.
  • Slab 3 included macrosynthetic fibers made of copolymer and polypropylene, up to 2-1/4 inches long, at a dosage of 7.5 pcy. There were no joints. Because the fibers used on slab 3 made the concrete stiffer, the water and superplasticizer dosage was increased.
  • Slab 4 contained no fibers, nor any other reinforcement. Crack-control joints were cut with an early-entry saw, 1-1/2 inches deep, on a grid of 15 by 17 feet. The east half of this slab was treated with colloidal silica and did not get the dissipating-resin curing compound used on the other four slabs.
  • Slab 5 contained no fibers, nor any other reinforcement. There were no joints.