Mark D. Whittemore and Peter E. Stamnas, New Hampshire Department of Transportation
Ten years ago, New Hampshire (NH) would have been considered an unlikely candidate to become involved in high performance concrete (HPC). NH’s subsequent success in the development and refinement of HPC and performance based specifications evolved through a series of three bridges. NH’s philosophy was to start simple and work towards more complex projects.
Route 104, Bristol
The first HPC bridge* was a straightforward 65-ft (19.8-m) single span structure. The concrete mix design for the bridge deck was provided in a prescriptive specification. A test placement was required. This gave the contractor and producer the opportunity to refine the mix and placement techniques. The required average concrete strength of 9400 psi (65 MPa) at 28 days for the girders was not as high as other states were targeting, but was still substantially more than NH’s typical design value of 5000 psi (35 MPa) at that time. The following lessons were learned on this bridge:
- The decision to start simple proved to be a wise one. Specifying higher concrete strengths would have accentuated the problems experienced in girder production. More testing and trial batching and stricter quality control were necessary.
- The post-bid meeting should be mandatory. Development of the HPC mix for the girders was time consuming; time to develop and test mixes must be considered in the project schedule. Cooperative involvement of all parties became a focus on future HPC projects.
- The air requirement of 5 to 8 percent was a significant reason that the higher girder concrete strengths were difficult to obtain. The percentage of air entrainment for the girders was adjusted down to a target value of 5 percent.
- Match curing for girder concrete was required and found to be very useful.
- The wider girder spacings of 12 ft 6 in. (3.81 m) resulted in the need for special formwork at an estimated increased cost of 75 percent over typical formwork.
- The emphasis placed on the importance of deck curing was very successful in achieving positive results. Eight years later, there are still no visible deck cracks.
- The biodegradable bags used for silica fume did not fully disintegrate in the mixing process. Blended cement with silica fume is the preferred method to include silica fume in the mix.
- Specify only what is needed. Research indicated that target values of 1000 coulombs for permeability could be increased to 1500 coulombs at 56 days.
Route 3A, Bristol
The second bridge* was a 60-ft (18.3-m) long simple span structure completed in The superstructure consists of precast concrete deck panels with a cast-in-place concrete deck overlay and New England bulb-tee (NEBT) HPC girders. Several new lessons were learned:
- The girder test section and the test deck placement helped placement during production proceed smoothly.
- The revised permeability limits and less stringent air requirements facilitated achieving the required girder strength.
- The match-curing system was very successful.
- The use of the deck panels was a much quicker and less expensive means of forming the deck.
The performance of Route 3A Bridge has been excellent. The only observed deck cracking consists of four short cracks.
Rollinsford Bridge
The third HPC project was a 110-ft (33.5-m) simple span bridge using NEBT girders. This was NH’s first project with alkali-silica reactivity (ASR) language in the specification. The addition of slag or Class F fly ash to control ASR introduced several noticeable changes in the mix characteristics.
Current Practice
NH has moved exclusively to a quality control/quality assurance (QC/QA) performance based specification for concrete bridge decks. The specification provides incentives for achieving consistency in results within given target ranges for concrete cover, air content, water-cementitious materials ratio, and permeability. A disincentive is provided for strength.
This specification has undergone numerous changes, most significantly in the permeability limits. With significant pay incentives for permeability, contractors pursued a high content of cementitious materials in the mixes. This practice resulted in substantial positive pay adjustments to the contractors but also much more deck cracking. The correlation between deck cracking and the high cementitious materials concrete convinced the NH Department of Transportation to increase the permeability target from 1500 to 2500 coulombs. The revised permeability specifications and more stringent curing specifications have led to a dramatic decrease in the number of concrete decks exhibiting cracking. The importance of proper curing cannot be overstated. Curing blankets (cotton mats or burlap) need to be on the deck and wet within a maximum of 30 minutes after concrete placement, and are required for a minimum of 7 days.
NH’s specifications continue to require girder test sections. Also, match curing of cylinders is now required for all prestressed concrete members. This benefits fabricators who are subsequently able to turn over their casting beds quickly. It provides the owner with the best non-destructive estimate of the actual concrete strength within the girder. Cylinder molds and controllers are expensive: ask only for the number of match-cured cylinders that you need.
The overall HPC philosophy has been incorporated into NH’s Standard Specifications and everyday practice. All of NH’s concrete decks have been QC/QA since 1999 and use of this performance specification will continue at least into the foreseeable future. Similarly, all concrete girders have used HPC since 2001. The successful construction and excellent performance of these structures have convinced NH of the benefits of HPC.
Editor’s Note
This article is the third in a series that describes how the use of HPC has progressed since it was first introduced into a State’s program. Other articles appear in Issue Nos. 30 and 35.
*See HPC Bridge Views, Issue Nos. 4 and 17.