Premature Rock Bolt Failure

This project was initiated in 1999 to address the observed phenomenon of premature failure of rock bolts in a number of Australian collieries, and with a particular focus on the problem of Stress Corrosion Cracking (SCC) in rock bolts. SCC occurs by the slow progressive growth of stress corrosion cracks under the joint action of stress and a corrosive environment. Eventually one of the cracks will reach a critical length at which the remaining section can no longer carry the load and final instantaneous overload failure then occurs. The stress corrosion cracks act as very sharp stress concentrators so that even small stress corrosion cracks can cause bolts to fail below their design strength, at times with catastrophic consequences.

The current project has surveyed the industry and found that the problem, although not identified in all mines, is more than just an isolated problem. It is not confined to just one coal seam where particular corrosive conditions apply or to all areas of that seam. It is not confined to one manufacturer or steel supplier. It is not restricted to old bolts ? some recently installed bolts have been found to fail under SCC conditions. It appears to be more prevalent where:

• Clay bands are intersected by the bolts
• Thick coal roof sections are present
• High tensile steel bolts are used
• There is some groundwater present in the strata
• There is limited shearing within the strata inducing bending in the bolts
• Bacterial ?bug? corrosion of steel underground may be active, promoting corrosion of existing flaws within the steel.

However, the data collected during this project does indicate that there are still significant problems in defining the scale of prematurely fractured bolts in coal mines. The causes of the uncertainty can be summarised as:

• Lack of a consistent, or formal reporting system for the identification of incidents of broken bolts.
• The results of any survey of premature bolt failures is limited by only identifying bolts that fail outside the encapsulated length of the bolt and are free to fall out or significantly displace out of the roof.
• To date the only means of identifying bolts that fail in the encapsulated section is through inspection of fall cavities.
• Even in areas of high broken bolt frequency no broken bolts were identified in high deformation roofs, with the exception of failure of the collar of the bolt ? bolts may be locked into the roof by shear action.
• Identification of the steel source is not possible without recourse to metallurgical analysis even for bolts with similar profiles.

There were at least four main types of premature bolt failure identified as part of this study. These were:

• traditional stress corrosion cracking failures caused by corrosion cracks or pits propagating until ultimate brittle failure occurs;
• existing cracks or pits in the steel which propagate and are the trigger mechanism for ultimate failure;
• brittle failures within the encapsulated section of the bolt without any visible corrosion;
• failures within the threaded section of the bolt typically across the root diameter of the thread.

Most failed bolts that were recovered from the mines were substantially straight, but this is considered to be simply caused by biased sampling of straight bolts being less likely to be locked into the roof. In fact, some shear loading mechanism is probably required to explain bolt failures, below the level of encapsulation, with very low tensile collar loads.

Fracture depth in bolts varied from <1mm to over 40% of the surface area of the bolt.

A series of metallurgical tests have been applied extensively to the sampled bolts to identify metallurgical characteristics of the failed bolts and the nature of the corrosion. These point to steels with a low fracture toughness as particularly susceptible to this type of problem, and related brittle fracture in the vicinity of the threaded section of the bolt (toughness is a measure of the energy required for fracture).

The latest developments of higher fracture toughness steels in Australia has produced variable fracture toughness results as measured by the Charpy test. This may be due in part to actual variability in the steels but also the inability of the Charpy test to adequately measure fracture toughness on quench and tempered or accelerated cooled steels. Alternative indicators of fracture toughness for rock bolts should be considered (eg bending tests , drop tests etc).

Anecdotal evidence and the latest use of higher fracture toughness steels by mines, indicates that higher fracture toughness steels are less susceptible to premature failure than steels with low fracture toughness. Tempcored steels and accelerated cooled steels in particular, appear to reduce the incidence of premature failures. However, the level of fracture toughness required to prevent premature bolt failures requires further investigation.

In addition to identifying that the problem of premature bolt failure due to SCC is occurring in a number of mines in both NSW and Qld, the current project has accessed a considerable database of similar problems from the UK coal industry over the past ten years or more. Furthermore, it appears that SCC is surfacing as a significant problem in a number of Australian metalliferous mines also. In their case, the prevalent use of point anchored bolts results in an even more severe problem due to 100% loss of capability on failure. A research project has recently been initiated in Western Australia into this problem in the metalliferous sector and agreement has been reached for collaboration through future information exchange.