have obviously put a great deal of effort into developing their classification system which differs considerably in detail if not in concept from those described by Abel 1, Wykeham et al 2 and B i eni a w s ki a, 4, 5. Unfortunately the system proposed by B a r t o net al. seems overcomplicated. As the authors themselves point out there are over 300,000 different geological combinations that can be represented. This number of possibilities is simply not "workable". The fact that this number is based on the analysis of 200 case records (commendable though that is) appears to be extrapolating to the extreme.These classification systems have definite limitations and it cannot possibly be expected that they will allow for all geological possibilities and stress conditions. The system described by B i eni a w ski a, 4, 5 recognizes this fact and is intended primarily for cases where tunnel wall displacements are the result only of rock mass loosening. Cases involving swelling rock/ gouge or extensive compressive failure of the rock mass (socalled "true" rock pressure) are specifically excluded. However, Barton et al. have attempted to cover all these possibilities. This appears to be asking too much of a classification system especially when the rock mass rating is related directly to tunnel support pressure. Particularly in the case of "true" rock pressure, many factors (e. g. excavation method, type and sequence of primary support installation), which cannot be accounted for in a classification system, will affect the load carried by the primary support G.Other drawbacks in the system proposed by B art o net al. appear to be:1. The only allowance for the strength of the rock material is in the SRF factor for "competent rock". For other categories of rock no account is taken of the material strenght. The term "competent rock" is itself not clearly defined.2. No allowance is made for the orientation of joint sets. This is unrealistic as it is well known v that the orientation of discontinuities relative to the tunnel axis can have an important influence on roof loosening.3. The system is based fundamentally on RQD which is both directionally dependent and rather sensitive to the skill of the driller.4. The system involves 9 rock mass classes and 38 categories for permanent support. These numbers are simply too large to be generally acceptable in practice. Similarly numbers such as 0.0015 or 0.018 for rock mass quality are likely to be viewed with some suspicion.

Summary --Zusammenfassung --R~sumdThe Shear Strength o~ Rock Joints in Theory and Practice. The paper describes an empirical law of friction for rock joints which can be used both for extrapolating and predicting shear strength data. The equation is based on three index parameters; the joint roughness coefficient JRC, the joint wall compressive strength JCS, and the residual friction angle ~r. All these index values can be measured in the laboratory. They can also be measured ~n the field. Index tests and subsequent shear box tests on more than 100 joint samples have demonstrated that ~r can be estimated to within _+ 1 ° for any one of the eight rock types investigated. The mean value of the peak shear strength angle (arctan T/~r~) for the same 100 joints was estimated to within 1/20 . The exceptionally close prediction of peak strength is made possible by performing self-weight (low stress) sliding tests on blocks with throughgoing joints. The total friction angle (arctan ~'/~r~) at which sliding occurs provides an estimate of the joint roughness coefficient JRC. The latter is constant over a range of effective normal stress of at least four orders of magnitude. However, it is found that both JRC and JCS reduce with increasing joint length. Increasing the length of joint therefore reduces not only the peak shear strength, but also the peak dilation angle and the peak shear stiffness. These important scale effects can be predicted at a fraction of the cost of performing large scale in situ direct shear tests. Die Indexzahlen k6nnen alle im Laboratorium bestimmt oder am Ort gemessen werden. Bestimmung yon Indexzahlen mit nachfolgenden Priifungen im Scherapparat von mehr als 100 Kluftproben haben erwiesen, daf~ ftir jede beliebige der acht untersuchten Gesteinsarten der Reibungswinkel ~r auf _+10 genau gesch~itzt werden kann.Rock Mechanics, Vol. 10/1-2 1 2 N. B a r t o n and V. C h o u b e y :Der Durchschnittswert des Reibungswinkels (arc tan (*lan) der Hdchstscherfestigkeit wurde fiir dieselben 100 Kliifte auf _+ 1/20 genau geschiitzt. Die besonders genaue Voraussch~itzung der Hdchstscherfestigkeit ist durch Eigengewicht-Gleitversuche (niedrige Spannungen) auf Gesteinsbl/Scken mit durchgehenden Trennfl~ichen erm/Sglicht. Der totale Reibungswinkel (arc tan v/an), bei dem das Gleiten eintrifft, ergibt eine Absch~itzung des Rauhigkeitskoeffizienten der Kluft JRC. Der Rauhigkeitskoeffizient bleibt iiber einen Normal-Spannungsbereich yon mindestens vier Grdgenanordnungen konstant. Die Indexzahlen JRC (Rauhigkeitskoeffizient) und JCS (Druckfestigkeitskoeffizient) reduzieren sich aber bei zunehmenden Kluftliingen. Bei zunehmender Kluftfl~ichengr~Si~e nehmen nicht nur die Hdchstscherfestigkeit, sondern auch der zugeh6rige Dilatanzwinkel und die Schubsteifigkeit ab. Diese wichtigen Einfliisse der geometrischen Abmessungen k~Snnen geschiitzt und zahlenm~if~ig erfagt werden, und zwar mit Kosten, die nur einen Bruchteil von denen betragen, die fiir grotge, direkte Scherversuche in situ erforderlich w~iren.

No abstract

No abstract

No abstract

No abstract

No abstract

No abstract