**3. Results and discussion**

### **3.1 Measurements in the experimental coal mine Barbara GIG**

The research work on the rock bolts grouted in a controlled way in real coal mine conditions yielded much information about the possibility of identification of grouting discontinuities and the influence of their changes on modal parameters. The measurements were performed in an experimental coal mine Barbara GIG. At the first stage the measurements were performed with previously developed working prototype assembled with National Instruments components, and at the second one, after large modification with the new measurement unit fulfilling ATEX requirements, shown in **Figure 8** (ATEX directives consists of two EU directives describing the minimum safety requirements of the workplace and equipment used in explosive atmosphere. ATEX derives its name from Equipment intended for use in EXplosive ATmospheres).

The total amount of investigated rockbolts in the experimental coal mine was 30. Initially, as it was not known how a supporting plate and a nut may influence the proper identification of grouting discontinuity the diagnose was realized on cases where supporting plate and nut were unscrewed and removed. Later on the experiments were performed with a complete assembly (a plate and a nut fixed), as shown in **Figure 9** and the results were compared (discussed in the second part of this paragraph). The characteristics of transfer functions for the analyzed cases

**Figure 8.**

*The portable measurement system for quality control of installed rock bolts, working prototype (left) and final version fulfilling ATEX requirements (right).*

were utilized for evaluation of natural frequencies, which were crucial parameters for proper matching with finite element modal models base and diagnosis of related discontinuity length. Basing on in situ measurements and the analysis, we concluded that damping did not convey satisfactory information on the subject and might vary to a certain degree from sample to sample overshadowing its proper usefulness. Since tests in real conditions were performed on a relatively short length of a rock bolt, a mode shape usage was also constrained.

The example results of the undertaken investigations are presented below. The comparison is made for measurements realized with the working prototype and the unit fulfilling ATEX requirements. The analyzed example case corresponds to the discontinuity length shown in **Figure 10**, the lack of grout from the drilled hole end and in the outer part of a rock bolt (supporting plate and nut unscrewed and removed). The differences in upper and lower plots may be attributed to different accelerometer orientation and consequently different impact direction. The identified natural frequencies are shown in **Table 4** and the scatter plots of the differences between FE model and data estimated experimentally for that case are presented in **Figure 11**.

In order to validate this method experiments were continued on the same cases of discontinuity deliberately prepared with the complete assembly of elements, so after screwing plate and nut to the rock bolt. Below the comparison of the measurements performed without the supporting plate and nut, and after screwing

### **Figure 9.** *Impact excitation of installed rock bolts, with the working prototype (left) and the unit fulfilling ATEX requirements (right).*

## *Quality Assessment of Installed Rock Bolts DOI: http://dx.doi.org/10.5772/intechopen.101125*

### **Figure 10.**

*The characteristics of the transfer functions for the analyzed case (measurements realized with the working prototype (a-upper chart) and the unit fulfilling ATEX requirements (a-lower chart)) and the estimated grout length (b), the length of the rock bolt is equal to 2.0 m. The chart axes are: the vertical axis—inertance, in (m/s<sup>2</sup> )/N, the horizontal axis the frequency, in Hz.*


### **Table 4.**

*The identified natural frequencies of the investigated rock bolt, measurement with the working prototype (a), and the unit fulfilling the ATEX requirements (b).*

them to the rock bolt is discussed. A typical torsion force applied in the real working conditions is 250 Nm, so such a value was used in the finite element model (FE). Of course it is not a constraint and other torsion forces may be used according to real situations; a thorough discussion on that topic is accessible in technical literature

**Figure 11.**

*Scatter plot of the differences between FE model and data evaluated experimentally. The values of the lengths of discontinuity are: (a) 19 cm and (b) 90 cm. These values are specified for the first minimum differences of models and are clearly outside the values of the random scatter. The upper plots are for measurements with the working prototype, the lower ones for measurements with the unit fulfilling the ATEX requirements.*

[20, 29]. The example cases (at first without a nut and a plate) are shown in **Figures 12** and **13**. Utilizing the characteristics of the transfer function for the analyzed cases (a), the scatter plots of the differences between FE model and data evaluated experimentally were used and the designated sections of the length of discontinuity were obtained (c). For the rock bolt grouted from a rear, hidden end, shown in **Figure 12**, the discontinuity length is approximately equal to 1.15 m. That value is specified for the first minimum difference of models and is clearly outside the values of the random scatter.

The experimentally identified and numerically calculated (FE model) natural frequencies are presented in **Table 5**.

For the rock bolt grouted from a roof strata surface, shown in **Figure 13**, the discontinuity length is approximately equal to 1.15 m and the length of the outer part is approximately equal to 0.16 m. These values are specified for the first minimum differences of models and are clearly outside the values of the random scatter.

The identified experimentally and calculated numerically (FE model) natural frequencies are presented in **Table 6**.

The results of measurements performed with the supporting plate and nut, after screwing them to the rock bolt are shown in **Figures 14** and **15**. For the rock bolt grouted from a rear, hidden end, shown in **Figure 14**, the discontinuity length is approximately equal to 0.95 m. Though there are two minimum values outside the random scatter, the first one may be chosen as valid, the second may be attributed to aliasing phenomena observed in frequency analysis as well.

The experimentally identified and numerically calculated (FE model) natural frequencies are presented in **Table 7**.

The grout length assessment is quite consistent with that obtained at the first stage, when a plate and a nut were unscrewed.

### *Quality Assessment of Installed Rock Bolts DOI: http://dx.doi.org/10.5772/intechopen.101125*

### **Figure 12.**

*The results of estimation of the grout length: (a) the characteristics of the transfer function for the analyzed case, the vertical axis of the chart—inertance, in (m/s<sup>2</sup> )/N, the horizontal axis—frequency, in Hz, (b) the scatter plot of the differences between FE model and data evaluated experimentally, (c) the estimated grout length, the length of the rock bolt is equal to 2.0 m.*


### **Table 5.**

*Comparison of identified natural frequencies for a rock bolt grouted from a rear, hidden end, the measurement using working prototype.*

### **Figure 13.**

*The results of estimation of the grout length: (a) the characteristics of the transfer function for the analyzed case, the vertical axis of the chart—inertance, in (m/s<sup>2</sup> )/N, the horizontal axis—frequency, in Hz, (b) the estimated grout length, the length of the rock bolt is equal to 2.0 m, (c) the scatter plot of the differences between FE model and data evaluated experimentally.*

For the rock bolt grouted from a roof strata surface, shown in **Figure 15**, the discontinuity length is approximately equal to 1.15 m.

The experimentally identified and numerically calculated (FE model) natural frequencies are presented in **Table 8**.

The grout length assessment is also quite consistent with that obtained at the first stage, when a plate and a nut were unscrewed.

### **3.2 Measurements in working coal mines**

Further experiments were realized in working coal and copper mines and around 50 rock bolts were tested. The aim of one of these experimental studies was connected with rock mass characterization [30] and examination of the strength of

## *Quality Assessment of Installed Rock Bolts DOI: http://dx.doi.org/10.5772/intechopen.101125*


### **Table 6.**

*The identified natural frequencies of the investigated rock bolt grouted from a roof strata surface, the measurement using unit using working prototype.*

### **Figure 14.**

*The results of estimation of the grout length: (a) the characteristics of the transfer function for the analyzed case, the vertical axis of the chart—inertance, in (m/s<sup>2</sup> )/N, the horizontal axis—frequency, in Hz, (b) the scatter plot of the differences between FE model and data evaluated experimentally, (c) the estimated grout length, the length of the rock bolt is equal to 2.0 m.*

rock bolts mounted in the rock strata [3, 4] at different depths. The study took place in a chosen corridor of the working coalmine. The rock bolts were grouted in the roof of the roadway. There were 12 rock bolts mounted in 4 rows and the lengths of the rock bolts were: 2.4 m, 1.85 m, 1.25 m and 0.85 m. Localization of the research and distribution of investigated rock bolts are presented in **Figure 16**. All rock bolts


### **Table 7.**

*Comparison of identified natural frequencies for a rock bolt grouted from a rear, hidden end for measurement units—the working prototype based on National Instrument's components and the new one fulfilling the ATEX requirements.*

### **Figure 15.**

*The results of estimation of the grout length: (a) the characteristics of the transfer function for the analyzed case, the vertical axis of the chart—inertance, in (m/s<sup>2</sup> )/N, the horizontal axis—frequency, in Hz, (b) the scatter plot of the differences between FE model and data evaluated experimentally, (c) the estimated grout length, the length of the rock bolt is equal to 2.0 m.*

*Quality Assessment of Installed Rock Bolts DOI: http://dx.doi.org/10.5772/intechopen.101125*


### **Table 8.**

*The identified natural frequencies of the investigated rock bolt, measurement unit—working prototype.*

### **Figure 16.**

*The exploited seam with investigated rockbolts and hydraulic jack for pull out test (a), distribution of measured rock bolts no 1–12 (b) and the example geometry of the identified discontinuity case (c). The lengths of rock bolts: no 1, 2, 3—2.4 m, no 4, 5, 6—1.85 m, no 7, 8, 9—1.25 m, no 10, 11, 12—0.85 m.*

were grouted using the resin material type Lokset. The grout length was 30 cm from the bottom of the hole.

Then a pull out test was conducted by technical staff, who made a thorough analysis of obtained characteristics of pulling (put forward) of the rock bolts taking into account not only pulling of the rock bolt from the grout but also extending of the rock bolt as a result of applied force.

The quality assessment of grouting of rock bolts was performed as complementary to these tests. Although localization and length of the grout were known, the research was undertaken assuming that the result of the grouting process does not necessarily coincide with the intended one. Following are the results of the identification studies of quality assessment of grouted rock bolts. The reference models (FE models), with a specific location of grout, corresponding to experimental cases were matched. The research was conducted for 7 cases and for 4 cases studies were performed before and after pull out tests. For the shortest rock bolts, length 85 cm, lack of sufficient grout strength was also observed (rocks were too weak at that lengths).

The examples of the analysis results in ANSYS environment are shown in **Figures 17** and **18** (visible parts are: a rock bolt and a resin layer). Correct matching cases with calculated mismatch errors are shown in **Tables 9**–**11**. The diagnosed grout lengths were localized at the end, bottom part of the rock bolts and were very close to the intended grout length of 30 cm. In order to check the accuracy of the assessment the FE calculations were performed also for smaller and larger grout lengths. For example, for the rock bolt with a length of 1.25 m, the smallest difference was obtained for the grout length 31 cm, and by increasing the length of the modeled grout by 1 cm the error changed from positive to negative values, which meant that the correct value was somewhere between 31 cm and 32 cm.

An important observation made during the tests was a slight but distinct increase of identified natural frequencies after pull out tests, which proves the

**Figure 17.** *The example of analysis results in ANSYS environment for a rock bolt length of 1.85 m.*

**Figure 18.** *The example of analysis results in ANSYS environment for a rock bolt length of 1.25 m.*

### *Quality Assessment of Installed Rock Bolts DOI: http://dx.doi.org/10.5772/intechopen.101125*


### **Table 9.**

*Rock bolt length 2.4 m, grout length 0.3 m from the bottom of the hole, case no 1.*


### **Table 10.**

*Rock bolt length 1.85 m, grout length 0.3 m from the bottom of the hole, case no 5.*

impact of the test on mechanical parameters of the test structure and shows a stress hysteresis. Because of the elongation, that is, a slight increase of the length of the rock bolt, this change should go in the opposite direction, namely a decrease of the natural frequencies. In analyzed cases it appears that the first factor is dominant—stress hysteresis. During a normal quality assessment of grouted rock bolts this effect will not take place. It should also be noted that in investigated cases not all natural frequencies were identified, however, their number was sufficient to match the experimental and theoretical (FE) models. The results were quite satisfactory and proved the usefulness of the method.

Based on obtained knowledge and experience research was continued for quality assessment of rock bolt support system realized as a project of above 2 km length corridor drilled for excavation purposes to enable access to large coal deposits. The


### **Table 11.**

*Rock bolt length 1.25 m, grout length 0.3 m from the bottom of the hole, case no 8.*


### **Table 12.**

*The identified natural frequencies of the investigated rock bolt.*

research was performed in several sessions and it seems to be relevant to perform such a control on a periodic basis.

The diagnosed natural frequencies with calculated mismatch errors for the example case are presented in **Table 12**.

The FRF functions (waterfall curve) for an investigated rock bolt, a placement of the response transducer, and the identified discontinuity length are presented in **Figure 19**. The discontinuity length is about 80 cm from the outer part. What was observed in that particular session that in the adjacent area several similar cases of discontinuity were diagnosed.

While explaining possible reasons of improper grouting it is worth considering the technology of installation of rock bolts. There are mainly three phases of fixing the rock bolt into rock strata: a placement of grout cartridges into a drilled hole using a rockbolt (a rock bolt is inserted up to it half length), turning phase with continued insertion of a rock bolt up to the end of the hole (depending on the environmental conditions a time period is about 10 s), spinning phase (about 4-5 s) and hold phase (about 15 s). It is very crucial to control these phases especially the turning and spinning ones. Otherwise lack of grouting connection may occur. Too fast insertion of a rock bolt may lead to leakage of grout from the hole and lack of grout in the back part. Too long spinning phase may cause damage of contact between a rock bolt and grout in the inner part (close to the end of the hole). It is

*Quality Assessment of Installed Rock Bolts DOI: http://dx.doi.org/10.5772/intechopen.101125*

**Figure 19.**

*The results of rock bolt quality assessment, FRF functions (waterfall curve) for an investigated rock bolt (a), the identified discontinuity case (b), a placement of the response transducer (c), the length of the rock bolt is equal to 2.5 m.*

because the hardening time in that part is shorter than that in the outer part (specific preparation of grout cartridges). Too slow insertion may result in not full mixing of grout, especially at the end of a rock bolt (one of the consequences might be a rock bolt sticking out more than it is supposed to). If there are crevices in rock strata some amount of grout may leak into that area with the result of local discontinuity of grout layer. Another case might be a larger diameter of the hole then projected, mainly in the outer part of a hole (quite often when rock strata is hard). Then the amount of grout inserted is not enough for proper connection of a rock bolt to rock strata and it also may lead to lack of grout in the outer part of the hole. Inspection of camera records of holes in the investigated area revealed that it could be the reason for not proper grouting for the case shown in **Figure 19**.
