Test Results on M16×80 10.9 Bolts with Varied Post Plating Hydrogen
Release Bake Times

The typical method for determining the efficacy of hydrogen release baking and thus the suseptability to hydrogen induced failure after release baking is to place the component under load for up to 48 hours and see if the component has failed due to stress related migration of the hydrogen to the highest stress locations. What has been developed is a nondestructive method that instantaneously determines the state of free hydrogen trapped in the substrate material trough observing dislocation mobility. There is a substantial amount of research work indicating that the presence of hydrogen alters dislocation behavior (Birnbaum, Robertson, Sofronis et al) and the increase in dislocation mobility is caused by hydrogen shielding of the dislocation sites. This phenomenon is observed and measured through insitu stress flow experiments of hydrogen charged metals under TE microscopy. This microstructure behavior is observed across both edge and screw dislocations in the microstructure of FCC, BCC and HCP ordered and disordered materials. EMAS is the only ultrasonic, NDT method capable of making measurements that determine the minute dislocation characteristics evidenced through the resonance attenuation behavior in the movement of hydrogen through the lattice.

The resonant frequencies of single dislocation lines are typically in the hundreds of megahertz. Low frequency dislocation damping models were developed by Mason, Truell and later Granato and Lucke and are now recognized as key theories in the field. Without repeating the mathematics derived for such
predictions, the principles for the (α) attenuation effects related to dislocation mobility are based on macroscopic variables of mechanical properties including shear modulus, etc. In this case data can be accumulated from various sample sets manufactured with common process parameters. Calibration values are retained and applied to subsequent bake performance measurements.

Presented are the results of the investigation of a method to determine the effect of different bake times performed at constant temperature and the effect upon dislocation mobility measured using electromagnetic acoustic spectroscopy. Dislocation mobility is assessed primarily through the attenuation behavior by measuring the rise and fall times of the resonance and anti-resonance peaks of low frequency (~10 MHz) resonances. The data indicates that the dislocation mobility is reduced or the hydrogen moved to trapping zones or released with increased bake times. It is critical that consistency in the measurement procedure is maintained and measurements taken at a specific point in the manufacturing cycle across a given manufacturing lot or sample set. The following graph is the data plot of the response times measured as the received signal pulse widths of the highest amplitude peak of a shear wave resonance detected at 8.12MHz.

Data Plot of Response Times

Zinc Plated, Medium Carbon Steel, baked at 200˚c, the Hydrogen Release Bake times of 0 through 8 hours. ASTM A574 equivalent. The graphed data are the
mean values of the sample sets provided by NORM Civata.

M16 Bolt Acoustic Response vs H Release Bake Time



This graph compares 2 bolts form the original set a 0 hour Bake (bolt #6) and an 8 hour bake (bolt # 3) both of these are baselines for the deplating tests. Here bolts were placed into a 15% Hydrochloric acid solution for 8 minutes to remove the zinc plating. As in pickling the HCL introduces Hydrogen into the samples, the amount of absorbed Hydrogen is not measured in this test. The data prebake on the No Bake, 0 hour Plated 224 microseconds and the deplated 0 hour, 216 microseconds and 8 hour, 210 microseconds bolts are within 7% in terms of the comparative acoustic response. The 8 hour bake, plated and 10 hour bake, deplated acoustic response times are within the data spread for the original 8 hour bake data set.

Method

The method applied is a narrowband continuous wave excitation system. Here a signal generator sweeps a narrow frequency bad determined by either
calculation or by empirically finding an appropriate frequency and performing comparative measurements. Driving a specially tuned bulk shear wave EMAT the complimentary receiving transducer simultaneously receives the resonant acoustic signal it is then demodulated and digitized. The signal is then computer analyzed for attenuation characteristics and Q. The technique requires that particular attention is paid to the direction of transducer orientation considering that most of the specimens under test have some anisotropy introduced at some point in the manufacturing process. Also for example in the event that irregular surface conditions the component may need to be rotated or positioned such that the signal peaks for a given resonance are maximized since the Q may be affected by typical variance in surface conditions. An important caveat here is in the event where components have been placed into service and are being tested for embrittlement damage the Rayleigh scattering will become the predominant indicator of micro-cracking and subsequent damage.

Interim Conclusion

Dislocation mobility as measured by electromagnetic acoustic spectroscopy is a high reliability method for assessing the efficacy of hydrogen release bake
efficacy with a small margin of uncertainty.