Experimental studies of the magneto-mechanical memory (MMM) technique using permanently installed magnetic sensor arrays

Abstract

The magneto-mechanical memory (MMM) method, that is often referred to as the metal magnetic memory method, has been reported to be a non-destructive testing technique capable of quantifying stress concentrations and detecting defects in ferromagnetic materials. The underlying mechanism behind MMM has been explained in the literature, but the sensitivity to stress concentration has not been satisfactorily investigated. In this paper, both the normal and tangential components of the stress-induced MMM signal were measured by permanently installed magnetic sensor arrays on specimens made from three grades of L80 alloy steel and 20 other structural steels; tests were also carried out on a pipe made from the 4140-L80 steel. As expected, the stress history affects the MMM signal, but the experimental results show that significant irreversible change of magnetization always occurs only in the first cycle of loading regardless whether the deformation is purely elastic or partially plastic. If the peak stress level is increased at a given point during cycling, the immediately following next cycle acts as a new “first” cycle at that peak stress level and causes additional significant irreversible change of magnetization, but there is no evidence that plastic deformation might build up a cumulative magnetization. The MMM effect is very small in the steel samples tested, indicating that it will not be useful in field applications. In un-notched specimens the irreversible change in magnetization caused a proportional change in the measured external magnetic field on the order of only 5–10 A/m, while in the case of notched specimens the leakage field was on the order of 30–60 A/m.