The Charpy impact test is one of the most common tests used to quantify the breaking strength of metallic materials. The test is implemented by striking a small, rectangular metal specimen with a large pendulum and recording the energy absorbed by the specimen as it breaks. Accurately determining the breaking strength of metals is critical in the construction of bridges, buildings, and pressure structures.
The low cost and simple configuration of the test have made it a common requirement in codes for critical structures. However, the test is not useful if the variation in test results between laboratories is not tightly controlled. The control is accomplished by requiring the indirect verification of impact machines. NIST administers a program to verify the performance of Charpy impact machines by selling specimens with certified breaking strength as a standard reference material (SRM). The program is conducted in accordance with ASTM Standard E 23-02a, "Standard Test Methods for Notched Bar Impact Testing of Metallic Materials".
The verification program works as follows. NIST obtains a pilot lot of 75 Charpy specimens from a supplier and then measures the breaking strength of the specimens using three master machines. If the measurements meet certain criteria, then the rest of the specimens are machined and sent to NIST. An additional 75 specimens are selected at random from the lot and broken. If the breaking strength of the additional specimens is in agreement with the pilot lot, then the lot is certified as a standard reference material by NIST. Customers test sets of five specimens (one SRM unit) and then return the broken specimens and observed values to NIST for analysis. The data are stored in a database for future reference. Charpy SRMs are the biggest selling SRMs at NIST, accounting for 22.4 % of all SRM revenue in FY09.
Statistical Engineering Division (SED) staff have been involved in the Charpy verification program since NIST acquired the program from the U.S. Army in 1990.
SED staff have proposed new verification limits for variation based on: an F-test, a range rule, and variation in the production lot, because the verification limits stated in E-23 are somewhat arbitrary and there is no limit to variation in a set of five specimens. We have developed an uncertainty procedure for the reference value as well as guidelines for computing uncertainty for customer measurements in the Recommended Practice Guide, "Computing Uncertainty for Charpy Impact Machine Test Results".
We are currently working to provide a new web-based proficiency testing service. SED staff are responsible for providing appropriate statistical analyses for the service.