Browse Projects > Detailed View

Cracking of Decks on Deep Plate and Tub Girders


Cracking of concrete decks supported by steel girders has been a perennial problem in the industry. Despite all research performed to date, this problem seems to be eluding a concrete (no pun intended) and satisfactory solution. The industry has seen a clear pattern of transverse deck cracks on several Tub and Plate Girder Bridges. There needs to be a solution.

The problem seems to be even more pronounced in the case of deep and stiff girders such as steel tub or deep steel plate girders. It seems the stiffer the girders, the bigger the problem. This research would propose to study deck cracking to identify at least the structural part of the problem. This would compare actual performance of bridge decks that have cracked with the analytical models to determine if we are capturing the actual stresses at hand. Other structural aspects to be researched would be:

  • Excessive shrinkage of the concrete
  • Curvature of structure
  • Stiffness differential: Deep steel girders are significantly stiffer than the deck
  • Thermal effects: Girder is expanding and deck is resisting the expansion
  • Effects of eccentricity: Girder is expanding due to temperature rise, the deck is resisting but at fibers that are further away from the neutral axis in these types of structures. Therefore, the stress they would see would be larger, potentially taking them above the modulus of rupture threshold.* Differential of the coefficient of thermal expansion: Typically, approximately 10%. New studies sponsored by Caltrans show the concrete coefficient of thermal expansion as low as 0.0000055 or lower, increasing the differential to approximately 12%.
  • Differentials in heat conductivity and ambiance differentials such as temperature and sun exposure: Steel is a very good heat conductor and can get heated faster and way beyond the temperature of the deck. Then the stiffness of the deep girders expanding would carry along the deck forcing it to crack for compatibility along the interface surface.
  • Longitudinal and Transverse Bending (for more flexible bridges), etc.* Placement, curing, and workmanship and specifications issues
  • Creep and shrinkage aspects.


The main objective is to understand the problem fully, the real causes of cracking, which might be singular but might also be a combination of all the mentioned basic cases. What causes a deck that has been designed to not see any tension in the deck from dead and live load to end up in the first few months of service with evenly spread cracks at every 2’-4’ frequency.

Once the problem is understood, it should be easier to mitigate the causes accordingly.

Research should be mostly analytical, combined with ambiance data from specific projects, but it may also involve some experimental testing/monitoring of new bridges.


Concrete decks in the USA crack. We have not figured it out yet as to what would be the right solution. Billions of dollars are spend replacing decks within 35-45 years, when their life should match that of the bridge to at least 75 years. Even when special mixes and protection are specified for the concrete mix, they usually assume concrete models with no cracks. That would indicate that even with special concrete mixes, and for bridges where decks have been designed for zero tensile stress based on the conventional service combinations of basic loads, the decks still crack and in frequencies that are similar from structure to structure, and are not easily explained from conventional practice of load combinations.

Related Research:

Research has been performed on the subject. It is believed that the structural portion is usually limited and the problem is assumed to be more linked to workmanship, and proper assurance of just bending behavior and assurance from the engineer that the concrete does not see any tension, at least from Dead Load and especially in the conventional positive moment region. Experience show that cracking in deep stiff girders does not distinguish between positive or negative moment regions, cracking is almost evenly distributed. So the problem should be also investigated as a structural problem linked to axial performance of the two materials and all resulting effects.


A combination of field measurements, lab testing and a lot of finite element studying of the issues noted above.


This issue could be implemented by all DOTs through bridge design manuals and specifications based on the research recommendations. This may also be implemented through specification changes by the AASHTO Committee on Bridges and Structures.


All bridge owners

Sponsoring Committee:AKB20, Steel Bridges
Research Period:24 - 36 months
Research Priority:High
RNS Developer:Tony F. Shkurti, PhD, PE, SE
Date Posted:02/14/2017
Date Modified:04/09/2020
Index Terms:Bridge decks, Girder bridges, Cracking, Stiffness, Shrinkage,
Cosponsoring Committees: 
Maintenance and Preservation
Bridges and other structures

Please click here if you wish to share information or are aware of any research underway that addresses issues in this research needs statement. The information may be helpful to the sponsoring committee in keeping the statement up-to-date.