Published Online: 1 December 2005
Page Count: 7
Professor, U.S. Naval Academy, Annapolis, MD
USNRC, Washington, DC,
NSWC Carderock Laboratory, West Bethesda, MD
(Received 1 April 2005; accepted 8 August 2005)
Tests conducted to identify the ductile to brittle transition in ferritic steels have historically been conducted at elevated loading rates since it was understood that this transition was very dependent on the loading rate. By testing at an elevated loading rate, the researcher identified an upper bound transition temperature that could be expected to be conservative for most structural applications. In the development of the Master Curve procedure and the To reference temperature of ASTM E 1921, Standard Test Method for Determination of Reference Temperature, To, for Ferritic Steels in the Transition Range, allowable test rates were restricted to the “quasi-static” regime typical of elastic-plastic fracture toughness standards like E 1290 and E 1820. Since this standard was developed primarily for nuclear pressure vessels in which even a pressurized thermal shock event results in relatively slow loading rates because of the large size of the structures involved, the limitation to rather slow loading rates was not considered to be important. The loading rate allowed by E 1921-03 encompasses approximately two orders of magnitude for dK/dt with dK/dt ≈ 1.0 Mpa √m/s, but the standard has not required the direct measurement of the loading rate or the reporting of the actual loading rate. The expectation is that for such a “quasi-static” loading rate the resulting To is not strongly dependent on the loading rate, and the result can be used in “quasi-static” applications without adjustment for the likely difference between test and application loading rates. Recently Hall and Yoon  and Wallin  have reported results that appear to show that tests conducted over the range of loading rates allowed by E 1921 can result in a difference in the resulting reference temperature of 30 to 50 °C. Wallin has suggested  reducing the allowable range of testing to a very narrow factor of 4 centered on dK/dt = 1.0 MPa√m/s, or requiring a correction procedure to account for the test machine rate used.
Based on the experimental work presented in this paper, and work of Hall and Yoon , a revision to E 1921 was recently approved by ASTM that reduces the allowable testing range from two orders of magnitude to effectively a range of 20, with a target test rate of dK/dt ≈ 1.0 MPa√m/s with the allowable range from dK/dt ≈ 0.1 MPa√m/s to dK/dt ≈ 2.0 MPa√m/s. Work is continuing in the E08.08.03 Task Group to develop a high loading rate procedure for E 1921, so that tests can be conducted at the rate of interest for the application.
In this work data are developed that shows that the E 1921-02 range is too generous, but it also implies that narrowing the present allow range by a factor of 5 from 100 to 20 is adequate to obtain a To measurement that is insensitive to the loading rate within 10°C.
Paper ID: JTE14108