Author Type

Graduate Student

Date of Award

1991

Document Type

Thesis

Publication Status

Version of Record

Submission Date

October 2026

Thesis/Dissertation Advisor [Chair]

Hartt, William H.

Abstract

A new experimental approach has been developed whereby fatigue crack growth rates are determined for multiple specimens tested in series. This involves a geometry whereby specimens respond to cyclic loading in a manner that is mutually independent. The classic compact-tension specimen geometry is inappropriate for this technique, since stress intensity factor range increases with increasing crack length (constant load amplitude or range); and this dictates load shedding according to the specimen for which crack propagation is most rapid. Stress intensity factor range for the remaining specimens (smaller crack length) consequently decreases to low values and may even fall to below Delta KI, such that little or no information is obtained from them. To overcome this difficulty a new constant taper, side-grooved compact-tension specimen geometry for which stress intensity factor decreases with increasing crack length at constant load has been developed. Load shedding is, therefore, not necessarily required to reduce Delta KI during the test. A finite element analysis was performed to establish the stress intensity-crack length (constant load), compliance-crack length and DCPD (Direct Current Potential Drop)-crack length character of the specimen. An appropriate side-groove depth to avoid crack deviation from the intended plane was determined experimentally, and an analytical expression to account for reduced thickness in the crack plane was verified. Calibration curves between crack length and compliance and between crack length and potential drop were verified as well, and the stress intensity-crack length characteristics for the specimen were confirmed by comparison of da/dN-Delta KI curves with those for conventional C(T) specimens. A locally fabricated, five specimen fatigue testing frame was developed in conjunction with a 98 kN MTS actuator and interfaced with a commercially available control system, the hardware and software of which were modified to accommodate multiple specimen control and data acquisition. The utility of the proposed specimen and test procedure are discussed within the context of threshold and near-threshold crack growth rate determination where long test times are normally required for data development, particularly for low cyclic frequency applications such as nuclear and offshore.

Extent

194 p.

Comments

College of Engineering and Computer Science|FAU Electronic Theses and Dissertations Collection|Thesis (Ph.D.)--Florida Atlantic University, 1991.|College of Engineering and Computer Science

Identifier

FADT12272

Format

pdf

Download

Share

COinS