Development of correlation between compressive strength and ultrasonic pulse velocity in rollercompacted concrete for dams

Abstract

Concrete is a fundamental building material that impacts construction productivity, time, cost, and structural durability. Dam construction projects, mainly using concrete, require careful consideration of time and cost factors when selecting an economically viable concrete type. Roller-compacted concrete (RCC) used to construct dams has become a modern choice, particularly for gravity dams, but the costeffectiveness depends on various factors. Concrete testing dates back to the 19th century and mainly focused on assessing concrete quality and longevity through compressive strength tests. Compressive strength tests are crucial for evaluating concrete properties using destructive and non-destructive methods. The U.S. Army Corps of Engineers (USACE, 1998) emphasizes laboratory investigation to characterize RCC properties, particularly in the context of the Elk Creek Dam. However, a research gap exists regarding the correlation between ultrasonic pulse velocity (UPV) and compressive strength in RCC dams, underlining the need for comprehensive testing methodologies to assist engineers in nondestructive testing of RCC strength. The study aimed to establish relationships between various compressive strengths of High-cementitious Roller-Compacted Concrete (HCRCC) used in dam construction by implementing both destructive (DT) and non-destructive (NDT) testing on laboratory HCRCC. After that, to establish a correlation between the DT and NDT testing results and validate the identified data through filed data. This research utilized two different High-cementitious Roller-Compacted Concrete (HCRCC) mix designs, varying in the total quantity of cementitious materials used. The two designs, 15/38-365 and 20/38-90 met the design requirements for Roller-Compacted Concrete (RCC). Applying a variable by replacing fly ash with cement for the total cementitious content resulted in eight distinct HCRCC mixtures. A total of 72 specimens were cured for 7, 14, and 28 days. The research conducted non-destructive testing (NDT) methods, conducting Ultrasonic Pulse Velocity (UPV) and Rebound Hammer (RH) tests for each specimen. Subsequently, following NDT, each specimen underwent a compressive strength test using the DT method. Field data collection occurred at the De Hoop Dam and Springs Grove Dam, chosen for their construction using Inverted Roller Compacted Concrete (IVRCC), ensuring an accurate reflection of measurements specific to RCC dam structures. A multiple regression model with a linear correlation was developed with the laboratory dataset without outliers, resulting in a strong positive model with an excellent R2 of 0,93. Expressing that 94% of the change in compressive strength was a function of non-destructive testing methods evaluated, with an error of 6% over the range evaluated. Further investigations are required for different RCC dam design mixes and the non-destructive testing thereof.