Browse Projects > Detailed View

Determining the Energy to Rupture of Crack-Like Flaws under Stress

Determining the Energy to Rupture of Crack-Like Flaws under Stress


Load-carrying structures like bridges contain flaws and defects that can eventually result in observable cracking. These flaws may be the result of welding defects, severe constraint and/or cracking during construction fit-up. The appearance of cracks and other defects noted by visual inspection is particularly problematic for non-redundant bridges that are deemed to be fracture critical. Because visible cracks can be indicative of the progressive nature of crack advancement under cyclic stress, they may result in eventual failure. If so, they must be addressed by repair, retrofit or even structural replacement. For bridges over features like major river crossings that carry hundreds of thousands of vehicles per day, delays caused by repair or replacement can be very costly in terms of construction, diversion of traffic, and loss of business.

Costs can range from several million dollars, up to one billion for replacement of a bridge over a major river if the extent of damage could result in complete closure. In the recent repair of the I-64 Sherman-Minton Bridge over the Ohio River between Indiana and Kentucky, the evaluation of potential failure was based on an assessment of whether crack-like flaws under the influence of residual stress from original fabrication could catastrophically rupture due to its ambient high dead-load stresses. The material involved in that problem was a proprietary high-strength quenched and tempered steel similar to ASTM A514. The problem is not specific to just to A514-type steels; it could also apply to A36, A588 or A572 or any of the A709 HP steels.

Research Need

The engineers who collaborated on the Sherman-Minton Bridge could not determine how the CVN test or KIc fracture toughness data correlated with crack-like defects that had semi-blunt notches at the leading edge of the flaw. The flaw notch tip had a severity that was less than a crack tip, but was not as blunt as the 0.010” notch radius used in the Charpy test. Moreover, although the flaws were in the weld joints, the residual stresses were not measured but estimated. The dead load stresses were known, but the residual stresses were not, nor were ways available to estimate what was the effect of this partially blunt notch at the crack terminus. Because it was unknown, fracture mechanics was used, which came to the conclusion that the bridge was in jeopardy, and was therefore repaired by use of retrofit plates. In fracture mechanics analyses, residual stress consideration may the controlling factor in determining the severity of cracks in structural steel members. But was the bridge in jeopardy?

Questions remain about the bridge because the subsequent failure analysis indicated that the crack-like flaws were probably related to shop welding, and they showed no signs of sub-critical fatigue crack growth after 50 years of service. Welding cracks generated during fabrication will begin as sharp cracks during their formation, but the crack tips may blunt themselves into something less than the sharp crack tips normally encountered in fatigue or stress-corrosion cracking. The fracture toughness of observable cracks having a notch tip severity less than a crack tip found in pre-cracked KIc specimens, but more severe than a Charpy notch, should be determined for common structural steels such as A36, A588, A572 and A514. This can be accomplished by using weld joints whose notch severity can be changed by the use of controlled root separations using either Charpy or dynamic tear specimens. This proposed work will reveal when does a sharp notch in a weld joint become a crack, and how much energy/unit area is required to form a sharp crack tip. In addition, the proposed testing will improve correlations with CVN and dynamic tear correlations with KIc. Additionally, the effect of preheat temperatures on residual stress in constrained weld joints needs to be determined. This can be very useful in estimating residual stress from a knowledge of preheat and interpass temperatures obtained from welding records of the fabricator.

This study can provide engineers better tools to evaluate the potential for failure when flaws of different lengths and notch tip severities are present in fracture-critical structures. It will also provide guidance on the magnitudes of residual stresses estimated from preheat and interpass temperature for common structural alloys such as A588, A572, and A514, and improve correlations between CVN and dynamic tear tests with KIc. The resulting information from this proposed testing will provide engineers with an improved basis for making repair or replacement decisions when cracks are discovered on steel bridge members in the field.

Submitted By:

Chris Hahin

Sponsoring Committee:AKC70, Fabrication and Inspection of Metal Structures
Date Posted:05/05/2013
Date Modified:05/07/2013
Index Terms:Bridges, Defects, Cracking, Bearing capacity, Failure, Residual stress, Retrofitting, Ohio River, Kentucky, Indiana,
Cosponsoring Committees: 
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.