Tunnel Support Design Using Rock Mass Rating
One of the most valuable applications of the Rock Mass Rating system is determining preliminary tunnel support requirements directly from the classification result. Bieniawski's 1989 support table provides empirical recommendations for rock bolt length and spacing, shotcrete thickness, and steel set requirements for each RMR rock class. This guide presents the complete support recommendation table, explains the stand-up time concept, discusses excavation sequencing by rock class, and works through a practical tunnelling example. For background on the RMR system itself, see our guide on what Rock Mass Rating is, and for more guides visit our geotechnical guides hub.
Why RMR is Used in Tunnel Engineering
Tunnel support design must address two fundamental challenges: preventing collapse of the excavation during construction and ensuring long-term stability throughout the operational life of the tunnel. Both challenges depend on understanding the mechanical properties of the rock mass surrounding the excavation. The RMR system provides a practical bridge between geological observations at the tunnel face and the engineering support measures needed to maintain stability.
The key advantage of RMR in tunnel engineering is immediacy. When the tunnel boring machine or drill-and-blast round exposes a new face, the geologist or engineer can assess the six RMR parameters within 15 to 30 minutes, compute the total rating, and determine the rock class. The support table then provides specific recommendations for that class, allowing support to be designed and installed before the stand-up time expires. This cycle of excavation, classification, and support installation repeats at every advance, making RMR an integral part of the construction management process.
RMR-based support recommendations have been validated against hundreds of tunnel case histories spanning a wide range of rock types, depths, and geographical settings. The 1989 support table represents the cumulative experience of the international tunnelling community, refined over 16 years of data collection since the original 1973 publication. While modern tunnel design increasingly incorporates numerical modeling, the RMR support table remains the standard first-pass assessment tool and serves as a benchmark against which more detailed analyses are compared.
Bieniawski 1989 Support Recommendations Table
The following table presents the support recommendations from the Bieniawski 1989 classification system for a horseshoe-shaped tunnel with approximately 10-meter span, excavated by drill-and-blast. These recommendations represent minimum support requirements and should be adjusted based on site-specific conditions, contractual requirements, and detailed design analysis.
| Rock Class | Excavation Method | Rock Bolts (20 mm dia., fully grouted) | Shotcrete | Steel Sets |
|---|---|---|---|---|
| Class I Very Good Rock RMR 81–100 |
Full face advance. 3 m advance. | Generally no support required except for occasional spot bolting. | None | None |
| Class II Good Rock RMR 61–80 |
Full face advance. 1–1.5 m advance. Complete support 20 m from face. | Locally, bolts in crown 3 m long, spaced 2.5 m, with occasional wire mesh. | 50 mm in crown where required. | None |
| Class III Fair Rock RMR 41–60 |
Top heading and bench. 1.5–3 m advance in top heading. Commence support after each blast. Complete support 10 m from face. | Systematic bolts 4 m long, spaced 1.5–2 m in crown and walls with wire mesh in crown. | 50–100 mm in crown and 30 mm in sides. | None |
| Class IV Poor Rock RMR 21–40 |
Top heading and bench. 1.0–1.5 m advance in top heading. Install support concurrently with excavation 10 m from face. | Systematic bolts 4–5 m long, spaced 1–1.5 m in crown and walls with wire mesh. | 100–150 mm in crown and 100 mm in sides. | Light to medium ribs spaced 1.5 m where required. |
| Class V Very Poor Rock RMR < 20 |
Multiple drifts. 0.5–1.5 m advance in top heading. Install support concurrently with excavation. Shotcrete as soon as possible after blasting. | Systematic bolts 5–6 m long, spaced 1–1.5 m in crown and walls with wire mesh. Bolt invert. | 150–200 mm in crown, 150 mm in sides, and 50 mm on face. | Medium to heavy ribs spaced 0.75 m with steel lagging and forepoling if required. Close invert. |
Several important points apply to this table. First, the recommendations assume a horseshoe cross-section. Circular tunnels (such as those excavated by tunnel boring machines) generally require less support than horseshoe sections of the same span because the circular shape distributes stress more uniformly. Second, the bolt lengths and spacings assume a 10-meter span. Larger spans require proportionally longer bolts and may require closer spacing. Third, these are minimum recommendations. Site-specific conditions such as high in-situ stress, water pressure, swelling ground, or seismic loading may require heavier support than indicated by the rock class alone.
Stand-Up Time and Excavation Span
Stand-up time is the length of time an underground excavation can remain unsupported after excavation without experiencing failure. It is a critical parameter for construction planning because it determines the maximum time available for installing support after each excavation advance. If support is not installed within the stand-up time, the unsupported rock may fail, potentially causing collapse, overbreak, and construction delays.
Bieniawski developed a relationship between RMR, excavation span, and stand-up time based on case history data from tunnel projects worldwide. The key values from this relationship are summarized below.
| Rock Class | RMR Range | Stand-Up Time | Unsupported Span |
|---|---|---|---|
| Class I — Very Good | 81 – 100 | 20 years | 15 m |
| Class II — Good | 61 – 80 | 1 year | 10 m |
| Class III — Fair | 41 – 60 | 1 week | 5 m |
| Class IV — Poor | 21 – 40 | 10 hours | 2.5 m |
| Class V — Very Poor | < 20 | 30 minutes | 1 m |
The stand-up time relationship demonstrates why excavation sequence and advance rate must change with rock class. In Class I rock, the tunnel can be driven full-face with 3-meter advances and support installed at leisure. In Class V rock, the stand-up time of 30 minutes for a 1-meter span means that support must be installed immediately after each short advance, and the unsupported span must be kept as small as possible. This fundamentally changes the construction method, advance rate, and associated cost.
It is important to note that stand-up time is not a precise prediction but an empirical estimate based on case history data. Actual stand-up time at a specific location depends on many factors not captured in the RMR classification, including the exact geometry of the excavation, the three-dimensional stress field, water pressure, and the presence of specific geological features such as faults or clay seams. The Bieniawski stand-up time values should be used as a guide for construction planning, not as a guarantee of stability.
Excavation Sequence by RMR Class
The excavation sequence describes how the tunnel cross-section is divided into smaller sections (headings) that are excavated and supported sequentially. The purpose of sequential excavation is to reduce the unsupported span at any given time, maintain stability of the temporary face, and allow support to be installed before the stand-up time expires.
In Class I and Class II rock (RMR above 60), full-face excavation is typically feasible. The entire cross-section is excavated in one pass, with advances of 1 to 3 meters per round. Support, if required, is installed behind the face at a distance of up to 20 meters. This is the fastest and most economical excavation method.
In Class III rock (RMR 41 to 60), the excavation is divided into a top heading and bench. The top heading (upper portion of the cross-section, typically 50 to 60% of the full height) is excavated first with 1.5 to 3 meter advances. Support is installed in the top heading before the bench (lower portion) is excavated. The bench is then removed under the protection of the already-supported crown. This sequence reduces the unsupported span in the crown where gravity-driven failures are most likely.
In Class IV rock (RMR 21 to 40), top heading and bench excavation continues but with shorter advances of 1 to 1.5 meters. Support must be installed concurrently with excavation, meaning that bolts and shotcrete are applied immediately after each blast round rather than at some distance behind the face. The shorter advance length keeps the unsupported area small enough to remain stable during the limited stand-up time.
In Class V rock (RMR less than 20), the cross-section may need to be divided into multiple smaller drifts, each excavated and supported independently before the adjacent drift is started. Advance lengths are reduced to 0.5 to 1.5 meters, and shotcrete is applied as soon as possible after blasting, often before the muck is fully removed from the face. Forepoling (installation of steel pipes or spiles ahead of the face) may be required to provide pre-support before excavation. In extreme cases, ground freezing, grouting, or pipe-roof umbrella techniques may be needed to stabilize the ground ahead of the face.
Shotcrete Design Considerations
Shotcrete (sprayed concrete) serves multiple functions in tunnel support: it seals the exposed rock surface to prevent loosening and weathering, provides a structural membrane that resists block movement, and distributes loads between rock bolts. The RMR support table specifies minimum shotcrete thicknesses for each rock class, but several design considerations affect the final specification.
Modern tunnel shotcrete is typically fiber-reinforced (steel or synthetic fibers) rather than plain shotcrete or mesh-reinforced shotcrete. Fiber reinforcement improves the post-crack ductility and energy absorption capacity of the shotcrete, which is critical for maintaining support function in deforming ground. The typical fiber dosage is 25 to 40 kg/m3 of steel fibers or 4 to 8 kg/m3 of macro-synthetic fibers. The 28-day compressive strength is typically specified at 25 to 40 MPa, with early-age strength requirements of 1 MPa at 8 hours for Class IV and V rock where rapid load-bearing capacity is essential.
Shotcrete thickness specified in the RMR table is the minimum design thickness, measured perpendicular to the rock surface. Actual applied thickness varies due to the roughness of the excavation surface. Overbreak (excavation beyond the design profile) increases the volume of shotcrete required and can create areas of excessive thickness that are prone to cracking during curing. Quality control of shotcrete application includes monitoring nozzle distance, water-cement ratio, fiber distribution, and actual thickness using probe holes or ground-penetrating radar.
Rock Bolt Design Considerations
Rock bolts in tunnel support function by reinforcing the rock mass and creating a self-supporting arch of bolted rock around the excavation. The RMR support table specifies bolt length, spacing, and diameter (20 mm fully grouted), but the design must also consider bolt type, grout properties, installation timing, and load verification.
Fully grouted rock bolts, as specified in the RMR table, are the standard support element for permanent tunnel support. They consist of a deformed steel bar (typically Grade 500 or Grade 600) inserted into a drilled hole and encapsulated with cement or resin grout. The grout transfers load between the bolt and the rock along the entire embedded length, providing distributed anchorage that is more reliable than point-anchored mechanical bolts in jointed rock.
Bolt length specified in the RMR table ranges from 3 meters for Class II rock to 5-6 meters for Class V rock. These lengths are appropriate for a 10-meter span tunnel. A common rule of thumb is that bolt length should be approximately one-third to one-half of the span for crown bolts and at least one-half of the wall height for wall bolts. For spans significantly different from 10 meters, bolt lengths should be adjusted accordingly.
Bolt spacing ranges from 2.5 meters for Class II to 1-1.5 meters for Class IV and V rock. The spacing determines the load each bolt must carry and the size of the rock block that could fall between bolts. Wire mesh is specified for Class III, IV, and V rock to retain small blocks and loose fragments between bolt heads. In modern practice, mesh is often replaced by fiber-reinforced shotcrete, which provides equivalent or better retention while also sealing the rock surface.
Installation timing is critical. In Class III rock, bolts should be installed after each blast round, typically within the same shift. In Class IV and V rock, bolts must be installed concurrently with excavation, often before the bench is removed. Friction bolts (Split Sets or Swellex) are frequently used as immediate temporary support because they can be installed rapidly without grouting, with permanent fully grouted bolts installed later to replace or supplement them.
Worked Tunnelling Example
Consider a 10-meter span horseshoe-shaped road tunnel being excavated through moderately jointed sandstone at a depth of 80 meters. The following field measurements have been recorded at the tunnel face.
Parameter 1 — UCS: Schmidt hammer testing gives a mean rebound number corresponding to a UCS of 75 MPa. This falls in the 50-100 MPa range, giving a rating of 7 points.
Parameter 2 — RQD: Core logging from a probe hole ahead of the face gives RQD = 68%. This falls in the 50-75% range, giving a rating of 13 points.
Parameter 3 — Spacing: Scanline mapping on the tunnel face shows a mean discontinuity spacing of 350 mm for the most closely spaced joint set. This falls in the 200-600 mm range, giving a rating of 10 points.
Parameter 4 — Condition of Discontinuities: Assessment of the five sub-parameters yields: persistence 1-3 m (4 points), aperture 0.1-1.0 mm (4 points), slightly rough (3 points), no infilling (6 points), slightly weathered walls (5 points). Total condition rating: 22 points.
Parameter 5 — Groundwater: The face is damp with minor seepage but no measurable flow. This corresponds to the "damp" condition, giving a rating of 10 points.
Basic RMR: 7 + 13 + 10 + 22 + 10 = 62
Parameter 6 — Orientation: The dominant joint set strikes perpendicular to the tunnel axis with a dip of 55 degrees in the direction of drive. This is classified as "favorable" for tunnelling, giving an adjustment of -2 points.
Adjusted RMR: 62 - 2 = 60
An RMR of 60 places this rock mass at the boundary between Class II (Good Rock) and Class III (Fair Rock). For design purposes, the conservative approach is to apply Class III support recommendations.
Support Design: Based on the Bieniawski 1989 table for Class III rock in a 10-meter span tunnel:
- Excavation method: Top heading and bench. Advance 1.5 to 3 meters in the top heading. Install support after each blast round. Complete support within 10 meters of the face.
- Rock bolts: Systematic 20 mm diameter fully grouted bolts, 4 meters long, spaced at 1.5 to 2 meters in the crown and walls. Install wire mesh in the crown. Given the RMR is at the upper end of Class III, a 2-meter spacing is reasonable.
- Shotcrete: 50 to 100 mm fiber-reinforced shotcrete in the crown. 30 mm shotcrete on the walls. Given the favorable groundwater conditions (damp only), 50 mm in the crown is adequate.
- Steel sets: Not required for Class III rock.
Stand-up time check: For Class III rock with a 5-meter unsupported span (top heading width), the estimated stand-up time is approximately 1 week. With 2-meter advances and support installed after each round, the actual unsupported duration will be a few hours at most, well within the available stand-up time. This confirms the excavation sequence is feasible.
This example illustrates how RMR classification translates directly into practical construction decisions. The geologist classifies the face in 15 to 20 minutes, and the support can be specified immediately based on the rock class. During construction, this process repeats at every advance, with the RMR record building a complete profile of ground conditions along the tunnel alignment.
Limitations
The Bieniawski 1989 support recommendations have several limitations that designers should keep in mind. First, the table was developed for conventional drill-and-blast tunnelling. TBM (tunnel boring machine) excavation produces a circular profile with less disturbance to the surrounding rock, typically requiring less support than the table indicates for the same rock class. Specific TBM support design methods should be used for mechanized tunnelling.
Second, the recommendations assume standard tunnel depths and stress conditions. At great depth (typically greater than 500 to 1000 meters depending on rock type), high in-situ stress can cause stress-induced failure modes such as rock bursting or squeezing that are not adequately captured by the basic RMR classification. In these conditions, supplementary analysis using stress-strength ratios and numerical modeling is essential.
Third, the support table does not account for dynamic loading from seismic events or blasting from adjacent excavations. In seismically active regions or near active mining operations, additional support capacity must be provided beyond the RMR table recommendations.
Fourth, the table provides support for a single tunnel opening. When tunnels are excavated in close proximity (such as twin-tube road tunnels), the interaction between the two excavations can increase support requirements, particularly in the pillar between them. Numerical analysis is recommended when the center-to-center spacing is less than approximately three tunnel diameters.
Finally, the RMR support recommendations are empirical guidelines derived from case history data, not first-principles engineering calculations. They should be used as a starting point for preliminary design and construction management. For final detailed design, the empirical recommendations should be verified and refined through numerical stress analysis, kinematic stability assessment of specific wedge geometries identified from discontinuity mapping, and instrumentation monitoring during construction.
Frequently Asked Questions
For a 10-meter span tunnel in RMR Class III (Fair Rock, RMR 41-60), Bieniawski 1989 recommends systematic rock bolts 4 meters long spaced at 1.5 to 2 meters in the crown and walls with wire mesh, combined with 50-100 mm of shotcrete in the crown and 30 mm on the walls. No steel sets are required for Class III. The excavation should advance as top heading and bench with 1.5 to 3 meter advances in the top heading, with support installed after each blast round and completed within 10 meters of the face.
Stand-up time is the duration an underground excavation can remain unsupported without experiencing failure. It depends on both the rock mass quality and the excavation span. Bieniawski's data shows that Class I rock (RMR 81-100) has a stand-up time of approximately 20 years for a 15-meter span, while Class V rock (RMR below 20) may have only 30 minutes for a 1-meter span. Stand-up time is a critical construction planning parameter because it determines how quickly support must be installed after each excavation advance.
The Bieniawski 1989 support table was calibrated for approximately 10-meter span horseshoe-shaped tunnels. For larger spans, support requirements increase: longer rock bolts, closer spacing, thicker shotcrete, and potentially heavier steel sets. For spans significantly larger than 15 meters, such as underground caverns or large mine openings, the RMR table should be used only as a preliminary guide, supplemented with numerical stress analysis and the Q-System support chart, which accounts for span explicitly through its Equivalent Dimension parameter.
Excavation becomes progressively more cautious as rock quality decreases. Class I and II rock (RMR above 60) allows full-face advance of 1 to 3 meters with support installed behind the face. Class III rock requires top heading and bench with 1.5 to 3 meter advances and support after each round. Class IV rock needs shorter 1 to 1.5 meter advances with concurrent support installation. Class V rock may require multiple drift excavation with 0.5 to 1.5 meter advances, immediate shotcrete application, and forepoling ahead of the face for pre-support.