Developed initially for coal mining, roadheaders first appeared on tunnel sites in the late 1960s and for a considerable period their domain was in the softer and friable rock. Growing stronger in power and heavier, they were soon also found in harder formations, mainly under restrictive environmental conditions – like shallow overburden and in the vicinity of sensitive structures. The shock-free excavation made them a useful piece of apparatus. These benefits, common to all means of mechanical excavation, were assisted by the versatility of the roadheaders’ size and shape – features common to drill and blast rigs. However, since the mid 1980s, roadheaders have increasingly faced competition from modified excavators and smooth blasting methods. As a consequence, roadheader excavation was partly shifted to niche markets.
The main disadvantages of roadheader excavation are that increasing rock strength and abrasiveness raise costs sharply, due to low performance and high pick wear. There is also increased risk in rock conditions that have not been, or could not be, investigated sufficiently. This risk is higher in tunnelling compared to mining, where any excavation method often has to cope with varying conditions. In such applications drill and blast excavation is less sensitive. Considering these facts, it was a matter of necessity to improve the capability of roadheader technology. The most important issue being to significantly improve their performance in hard rock.
Demands on roadheaders in harder rock
In recognition of both the existing limits and the prospective demands on roadheaders in harder rock, several targets were identified to form the basis of a development programme:
In order to achieve these goals, an improved understanding of the interaction between rock behaviour and the cutting process was needed. The introduction of roadheaders with low cutting speeds, in order to keep the machine and (mainly) its picks within tolerable operating costs in hard and/or abrasive rock, was necessary. This needed to be combined with the utilisation of higher torque ratios, with respectively higher pick forces, in order to increase the cutting rate and to shift rock strength limitations. The system and cutting tools would then need to be developed and tested, and this research backed up with theoretical evaluations, lab tests and final verification in instrumented field tests.
Rock tool interaction
Investigations into rock tool interaction were initiated in the late 1980s, when a full scale test rig was installed at Voest Alpine’s Zeltweg facilities. The target was to close the gap between standard rock tests in a lab and site measurements. Measuring pick forces involved measuring energy and pick wear in different rock types under varying cutting parameters and provided improved understanding of the rock failure processes under a pick. Even more importantly, it provided a rational and quantifiable base for the design of cutterheads and pick lacing.
One of the first successful results with this ‘new’ generation of cutterheads was achieved in Spain. In the Balsareny potash mine the newly introduced 91t, Alpine Miner AM85P, with 270kW of cutter-power, out performed its predecessor, the AM100 (105t, 300kW) to a considerable extent. The net cutting rate was increased by more than 30% and monthly production by almost 60%.
Low speed cutting
The influence of cutting speed on the cutter life has been known since the late 1960s, when Russian (Krapivin et al., 1967) and German (Schimazek, 1970) investigations first indicated the strong correlation between rock type, rock strength and abrasiveness, cutting speed and pick consumption (figure 2). The main reasons for this effect were found in the relation between heat generation and cutting speed, leading to weakening and thermal failure of temperature sensitive tungsten carbide. Transition from abrasive wear to adhesive wear also contributes to this process. The impact load on picks rises over-proportionally with cutting speed and rock strength too, resulting in loads exceeding the mechanical strength of a pick tip.
These experiences soon led to the application of low cutting speed (<2m/s). Two-speed machines applying pole-changing cutter motors helped improve pick life in harder and abrasive formations, but as a consequence, there was a drop in power to the cutterhead with the ratio of speed reduction. With the introduction of the Alpine Miner AM105, speed reduction was solved for the first time by a high to low speed gear switch. Now the full cutter head power of 300kW was also available at low cutting speed.
At the Metro Pusan in Korea, an underground ridge of little altered granite (100MPa-170MPa, CAI of 2.7-4.3) had to be passed through. Since a hospital was situated over the top of the excavation area, blasting was not an option. So, for the first time, the ATM105 proved its capability in tunnelling through hard rock. The average cutting rate of 35m³/h was certainly a good result, but pick consumption (>5 picks/m³) and instability of the machine, strongly confirmed its limits.
During early 1996, Voest Alpine delivered another ATM105 for the rock section of Sydney`s Southern Rail Link. The rock encountered, a coarse grained sandstone with high quartz content, was not very hard (UCS 20MPa-70MPa) but its high abrasivity (CAI 2.7-3.6) and high fracture toughness made mechanical cutting difficult – as a number of roadheader operations in Sydney had shown before. Under these conditions the ATM 105 achieved very impressive results, with a net cutting rate of more than 100m³/h with a pick consumption of around 0.02 picks/m³ (Greimel & Gehring, T&TI, March 1999).
Comparing the results of Pusan and Sydney the following conclusions could be drawn: as long as the rock does not exceed a certain strength, existing pick technology would withstand the higher loads, even in highly abrasive rock. The same goes for the machine. The most important factor identified, compared to hard rock cutting, is the lower contact pressure between picks and rock. Thus, no overcritical heat generation, even in highly abrasive rock, was evident. The facts in hard rock application are quite different and therefore could not be mastered with low speed cutting alone, although it provided the most important input. The main differences to soft rock cutting result from the higher forces involved, which lead to:
These differences defined the requirements to be addressed in the ICUTROC project.
The ICUTROC R&D project
The complexity of the research undertakings called for an adequate organisational framework. Therefore, a corporate research project with partial funding from the European Community was established. Besides Voest Alpine Bergtechnik, it incorporated Sandvik Rock Tools as a major pick producer, research institutions and companies from the mining and construction industry. This team provided a firm theoretical background and a close connection to practical industry demands. The results of the actual R&D project have already been reported in detail (Akerman et al., 1998; Lammer & Gehring, 1999; Gehring & Reumueller, 2002). Therefore only the most important steps and results shall be presented here in a brief summary.
A new innovative grade of tungsten carbide (S-grade) was developed and patented by Sandvik Rock Tools in 1996. This grade provides highly increased mechanical and thermal properties. Comprehensive lab tests and investigations of wear mechanisms and thermophysical behaviour manifested the superiority of this new grade, this was also verified in comparative underground tests.
Because this new carbide behaves similarly (although less significantly) to other mine-grade carbides when exposed to extended excess temperatures (softening, thermal cracks), a new pick cooling system was designed. Each pick is supplied with water, defined according to cooling demand, with the intermittent spray directed on to the carbide tip. A 25% drop in the amount of water required, was a welcome side effect.
With a pre-requisite to utilise higher pick forces to their full potential and considering the characteristics of the machine, along with hard rock properties expected, a new generation of cutterheads with a reduced number of picks were developed. The force per single pick could be maintained within a non-critical range and the specific track length could be greatly reduced.
Substantial improvements have also been achieved in stiffening the machine, by using hydraulic boom stabilisation. Larger dimensioned boom cylinders and the introduction of pre-stressed vertical cylinders provided increased boom stiffness, maintaining the pre-set cutting depth (higher cutting rate) and sharply reducing vibrations (less ‘bumping’ of the boom).
Higher induced forces and increased system stiffness resulted in higher reactive forces being absorbed by the machine. A FE-model of the complete roadheader was set up and loaded with different configurations. The loads on the most stressed parts of the system like the turret, cutter gear box and frame were identified and reinforced accordingly.
Field tests
From the outset of the ICUTROC project, confirmation of project targets by field tests was seen as an integral part of the entire task. This was not only to prove the interaction between individual developments and their contribution to the objectives, but also to obtain field acceptance. Two steps of testing were planned. Firstly, a fully instrumented test in an underground site was to be conducted, in order to verify the results and compare them with an existing AM105 with same power and comparable weight. Secondly, the new machine was to be tested under normal, everyday operating conditions, both in a mine and on a construction site.
The first, instrumented, field test was executed in the Erzberg iron ore mine in Austria. There the Mining University in Leoben runs a test mine and the prevailing rock was well within the project targets. Besides confirming the design and development, the results also showed a substantial improvement of machine capabilities (Table 1). The test operations in a German coal mine did not provide very impressive results. The main reason for this was that the sandstone, which was expected in the face after a few meters, did not show up. Thus no real hard rock cutting could be exercised.
The first trial application on a tunnelling project took place in the Pozzano Tunnel, south of Naples in Italy. Here, an almost finished road tunnel offered a length of 60m, where it crossed an old railway tunnel with 15m clearance. Blasting was not an option, as vibrations were limited to a maximum of 1.8mm/s. The rock encountered was a thickly bedded and partly silicified dolomitic limestone with a UCS averaging ~190MPa. Higher values and high abrasivity reflect silicified layers
Both rock strength and abrasiveness provided a challenging task. Nevertheless, the results, as shown in Table 2, were very impressive and considerably exceeded targets (Pichler & Gehring, 2000).
The American College Tunnel, in Greece, is part of the new expressway ring bypassing the north of Athens. The rock formation encountered was a sequence of crystalline limestone with minor carbonatic schist and phyllite. Interbedded in all rock types was a significant amount of quartz lenses. Due to the performance of the roadheader the contractor ordered a second machine in June 2001, both machines broke through in December 2001. One of these machines will also start excavation in a tunnel on the Athens’s Metro extension in 2002.
The task for the AM105-IC on the Premadio Power Station in Italy, was the excavation of the TBM start chamber and the extension of an existing powerhouse. Power generation had to continue during excavation without any interruption. The existing cavern was lined with a segment system very sensitive to any kind of vibration. For these reasons both client and contractor opted for roadheader excavation.
The project is located in a metamorphic series comprising gneiss, mica schist, chloritic phyllite, and a considerable content of quartz lenses. The distribution of individual rock types is highly changing with gneiss predominating. High strength occurs only in the gneisic and the quartz portion, but the average level of abrasiveness is generally high (Table 4). The first job for the ATM105-IC was the excavation of an assembly chamber for the TBM. This chamber had a trapezoidal section of 30m²-45m² and a length of 130m. Here the roadheader averaged at a net cutting rate of 90m³/h and a specific pick consumption of 0.21 picks/m³.
The next task was the extension of the powerhouse. A section of 31m width and 30m height resulted in a total volume of approximately 22,000m³ being excavated in levels of 3m each. The restricted dimensions and efforts to stabilise the wide span section did not allow high utilisation, but the AM105-IC successfully completed excavation in June 2002.
In the Bileca Headrace Tunnel, in Bosnia Herzegowina, another ICUTROC roadheader has been in use since June 2002. Approximately 3km of the 15.7km long tunnel (35m² section) is designated for roadheader excavation, the rest for TBM. According to pre-investigation the tunnel was expected to be in bedded cretaceous limestones including single dolomite layers (60MPa-100MPa with peaks of up to 150MPa). Meanwhile the face is located in hard and massive limestone (150MPa->200MPa). Although the operation is still in its ‘learning period’ the machine is already achieving an average cutting rate of ~30m³/h.
Influence of rock mass features
An important side effect of low speed cutting, detected in the course of the ICUTROC research, is that the contribution of the rock mass features to the excavation process increases significantly with decreasing cutting speed. Once this influence was evident, field investigations were executed at more than 70 different sites with different roadheaders (cutting speed, power and weight). For the evaluation of excavation-related rock mass behaviour the RMR rating (Bieniawski) was revised (RMRrev). The ratio field measured net cutting rate (NCReff) and theoretical net cutting rate (NCRtheor) was used to quantify the improvement of performance. A detailed report about this investigation was presented at the ITA conference this year (Gehring & Fuchs, 2002), however, the main results are summarised in Figure 4.
Conclusion
To date, 16 ICUTROC developed roadheaders have been delivered, five of these to tunnelling projects. All of them running in conditions exceeding the previous range of economic roadheaders. This proven capability calls for improved utilisation (improved logistical tuning of potential sites) in order to exploit the increased the range of application. A sufficient volume of roadheader related pre-investigation and qualified interpretation of data therefore still needs to aid this task.
Related Files
Fig 2 – Influence of cutting speed on pick wear (Krapivin et al. 1967)
Main interrelations of demands and how they were covered in the course of the ICUTROC project
Fig 1 – Targets of hard rock roadheader development
Fig 3 – Influence of rock mass features on cutting rate (comparison: low speed/high speed)
