**Abstract**

The chapter presents a method for non-destructive identification of discontinuity of a resin layer (grout) surrounding rock bolts. The method uses modal analysis procedures and is based on an impact excitation where a response transducer is positioned at a visible part of a rock bolt. Since the installed rock bolt acts as an oscillator, its modal parameters are changed by different lengths and positions of grouting discontinuity. Thanks to proper extraction of these parameters, with a resonant frequency seen as the most valuable, the intended identification is possible. The measurements and analyses were performed in laboratory conditions and subsequently at experimental and working coal mines where the measurement system was verified. The developed finite element model of the system under test, rock bolt - resin - rock mass, may be used as reference data base for investigated rock bolts. The advantages of the method include plausibility of grouting discontinuity assessment at any time after its installation, a nondestructive character of the method and the fact that it is not necessary to install any additional equipment into a roof section. It enables a localization of a grout discontinuity, whether it is the back part or the front part of a rock bolt.

**Keywords:** rock, rock bolt, safety, modal analysis, mining

### **1. Introduction**

The role of a rock bolt support system is to secure and reinforce the rock zone in the near field of an underground opening and to fasten it to deeper rock strata [1, 2]. Mostly steel rock bolts are used for that purpose [3, 4]. The rock bolt consists of a steel bar grouted in an oversize hole. A portable installation machine is used to spin the bolt into the hole filled with fast setting epoxy resin cartridges. After hardening of the resin layer a plate and nut are driven up the bolt. Although robust resin cartridges are used, in mining practice the rock bolt may not be fully encapsulated as a consequence of various geotechnical conditions [5–7]: rock divergence, escape of grout into crevices, rock strata movement and improper grouting. The lack of proper grouting may be very hazardous and should be monitored [8]. Current publications in the field of rock bolt diagnosis, indicate that much effort is being taken to estimate the rock bolt integrity and grout quality in the most precise way. Different approaches have been proposed e.g. Granit, Boltometer, RBT and other inventions or methods [9–14]. These methods rely on excitation of a tested rock bolt to vibration along its axis of symmetry and the analysis of output signals. Depending on the proposed method, both acoustic and ultrasonic waves are

generated. Correspondingly, different analytical approaches are used as wavelet transform analysis, Fast Fourier Transform (FFT) and neural network algorithms [5, 14–18]. Also smart sensors techniques are introduced for observation of behavior of grouted rock bolts [19], in particular load measurements at the head of them [20], but the problem is not yet fully resolved.

Accordingly, the method for non-destructive identification of grouting discontinuity of rock bolts is proposed to extend the diagnose scope in rock bolting. Thus the diagnosis of void spaces—regions of lack of bonding is seen as crucial here. In the method a transverse excitation is applied which is seen as more adequate for that purpose. At present a diagnose is completed after the analytical phase has been performed in laboratory conditions, so results are not accessible in situ. Its usage is restricted to steel rock bolts up to 2.5 m long, though not only in mining but also in building engineering. It is worth noticing that the same approach to test the integrity of installed rock bolts was described by Godfrey [21]. Though not known to the author at the experimental and analytical stage of the current work, it is very encouraging that similar methodology was presented over forty years ago.

The chapter starts with a theoretical description of the main rules and relationships between investigated modal parameters and measured data. Then the method and structure of a reference base of FE models are presented. Subsequently it is shown how the method was validated in an experimental coal mine. Finally the results of the research realized in working coal mines are discussed.
