ARA
ARA

 

 

Acoustic Emission Testing of Unidirectional Composite Coupons

Problem, Statement, and Objective

Many composite materials experience time-dependent microstructural damage at loads far below their failure stress. Damage here refers to ply-level microstructural changes, such as matrix cracking and fiber-matrix debonding. Some examples of this damage are shown in Figure 1. Such microcracking can have a significant softening effect in the off-axis fiber directions and cause more serious forms of damage, such as transverse cracking, delamination, and fiber failure. Due to the scale of this microcracking, the only practical method of monitoring its progression in response to various load histories is by the Acoustic Emission Method. We have developed an experimental and data analysis method for tracking these types of damage real-time.

Microcrack growth was monitored for three combinations of ply-level stress by testing 90, 45, and 30 degree off-axis coupons with the sample configuration shown in Figure 2. Due to changes in acoustic wavespeed with fiber direction, the relative arrival time of an acoustic event emitted from a microcrack depends on its location relative to the centerline of the coupon, as shown in Figure 3. By using a sensible acceptance criteria, only events from the shaded portion of the sample are accepted. In this way, only events from the uniformly stressed portion of the material, and not from other extraneous sources, are counted as "good events" (those events indicating the dynamic growth of a microstructural crack).
Figure 4 shows a location histogram of microcracking in the free length of a 90 degree sample, indicating the microcracking is uniformly distributed throughout the material. Each microcrack initiates at some stress level, depending on the load history, runs dynamically and stops due to some crack inhibitor (fiber, low stress region, etc.). Cumulative AE events detected in 90 samples are plotted versus axial stress in Figure 5. Notice that the onset of nonlinearity in the stress-strain response was found to correlate with the onset of significant AE. Also notice that the cumulative number of events show no pattern with loading rate. Apparently, there is significant scatter between samples in terms of the number of detectable events. This scatter completely masks any rate-effects present. However, the cumulative distribution function (CDF) of detected microcracking yields the same distribution with load level, as shown in Figure 6. The rate-dependence of microcracking in 90 degree samples is therefore weak. Use of the CDF simplifies analysis of data where the "detectability" of events may be an issue (e.g. data comparison from sample to sample and for different sensor types or different materials).

By tracking the energy of the AE waveforms, important characteristics of the microcracking can be inferred. Larger cracks will emit stronger acoustic signals as a larger amount of material has failed. Figure 7 shows the average event energy versus stress level from a 90 and 45 degree sample. Except for just prior to failure, average event energy, and therefore average arrested crack length, is found to be independent of stress level. The events which did display an increase in energy are isolated to a particular region of the material as shown in Figure 4(b).

Reference

  • Bocchieri, R.T., R.A. Schapery and M. Gorman, "Time-dependent Microcracking Detected in a Rubber-Toughened Carbon/Epoxy Composite by the Modal Acoustic Emission Method," Journal of Composite Materials, Vol. 37, No. 5, 2003, pp. 421-451.
    Abstract
    AE

For inquiries or comments, please contact:
Dr. Steven Kirkpatrick
Principal Engineer
e-mail: skirkpatrick@ara.com

Dr. Robert T. Bocchieri
Principal Engineer
e-mail: rbocchieri@ara.com

 

 

ae1

Figure 1.Microstructural Damage Void with Microcrack and a Fiber/Matrix Debond.

speciman

Figure 2. Specimen Configuration

events

Figure 3. Acceptable AE Events

histogram

Figure 4. Location Histogram of Microcracking

event2

Figure 5. Cumulative AE Events

distribution

Figure 6. Cumulative Distribution Function of Microcracking

energy

Figure 7. Average Event Energy