The conventional way to assess drilling performance in the oil field is to compare actual performance to statistical standards derived from offset records. By their nature, these standards are subjective and variable. While they are ideal to monitor short-range performance and trends in well-known, older fields, they lack the power of physical models to establish absolute, technical performance standards. Several authors have proposed models in which the drilling process is treated as a mechanical, energy-balanced system. Mechanical specific energy input, drilling efficiency, and a minimum specific energy equal to the rock strength are the three key elements of these systems. Full-scale simulator tests were conducted under controlled laboratory conditions to develop and validate an energy-balanced model for drilling under hydrostatic pressure. In these tests, wide variations in drilling efficiency from a few percent to three and four times were observed in the spectrum of common drilling practices. They provided new insights into the drilling process and the nature of common drilling problems. Analysis of field data revealed good correlation between simulator and field results. Using mechanical efficiency, specific energy input, and a bit-specific coefficient of sliding friction as key indexes of drilling performance makes bit selection and the diagnosis of failures and drilling practices more accurate and less ambiguous. Introduction Teale derives the following equation for specific energy in rotary drilling:He also introduces the concept of minimum specific energy and/or maximum mechanical efficiency. The minimum specific energy is reached when the specific energy approaches, or is roughly equal to, the compressive strength of the rock being drilled, i.e.,The mechanical efficiency is thenand the maximum efficiency is reached whenTeale recognized that the specific energy cannot be represented by a single, accurate number since the drilling process is characterized by wide fluctuations of the drilling variables due to its complex dynamics and the heterogeneous nature of rock. Approximate and mean values were found to be sufficiently accurate for a model to predict and analyze drilling performance. P. 373^

This paper describes a new approach to understand the causes of stick-slip vibrations experienced by PDC bits. This new model takes into consideration not only the axial and torsional modes of vibration, but also the coupling between these two modes through bit-rock interaction laws, which are consistent with laboratory results from single cutter experiments. These interaction laws, which account for both frictional contact and cutting processes at the bit-rock interface, are formulated in terms of the depth of cut, a variable that brings into the equations the position of the bit at a previous a priori unknown time. They also account for potential loss of frictional contact between the wear-flats and the rock. The delayed and coupled nature of this interaction is ultimately responsible for the occurrence of self-excited vibrations, which can degenerate into stick-slip oscillations under certain conditions. The features of the torsional vibrations that are predicted with this model are well in accordance with field observations. Furthermore, the results reveal that the model predicts apparent rate effects as an inherent outcome of the nature of the bit-rock interface. Introduction Different types of vibrations can take place while drilling with PDC bits, namely lateral (or bending), vertical, and torsional vibrations. The research reported in this paper focuses on torsional vibrations and more precisely on their most violent extent: the so-called stick-slip vibrations. Stick-slip, as indicated by its name, is characterized by a periodic (or almost periodic) succession of sticking phases where the bit stops for a finite time interval (typically several tenths of a second) and slipping phases where its angular velocity O increases up to two or three times the imposed velocity Oo, see Figure 1. Stick-slip oscillations can last for several minutes. Downhole measurements indicate that the magnitude of vibrations (in terms of torque and angular velocity) occurring close to the bit are up to one order of magnitude larger than the vibrations recorded at the surface (Cunningham, 1968; Lesseultre and Lamine 1998). The vibrations induced in the drilling assembly while drilling with PDC bits increase bit wear, and can result in premature fatigue failure of the string or breakage of the bit itself. According to a study performed at Elf-Aquitaine by Henneuse (1992), the occurrence of stick-slip corresponds to about 50% of the on-bottom drilling time. Field and Laboratory Observations The measured frequency fo of torsional and stick-slip oscillations is found to be close to the estimated first natural frequency of the drill string assembly in torsion (Brett 1992, Pavone and Desplans 1994, Challamel 2000). However, while the angular velocity exhibits a monofrequential behavior during stick-slip oscillations, the weight-on-bit W and the torque T present a more widely spread power spectrum, see Figure 2. Stick-slip vibrations have been observed to occur mostly under large weight-on-bit and low angular velocity (Brett 1992). Thus, an increase of the applied weight-on-bit or a decrease of the angular velocity can trigger instabilities leading eventually to stick-slip vibrations, see Figure 3. Similarly, a decrease of the applied Wo or an increase of Oo can stop stick-slip vibrations as shown in Figure 4. As depicted in these figures, stick-slip oscillations are accompanied by a decrease of the mean side (or lateral) acceleration. These data have been recorded downhole using the instrumented bit developed by Security-DBS (Lesseultre and Lamine 1998). Pragmatically in the field, it is sometimes necessary to pull the bit off-bottom in order to arrest the torsional vibrations (Brett 1992).

Surface mud logging systems have detected a severe form of torsional vibration that causes rapid destruction of polycrystalline diamond compact (PDC) drill bits. These failures can occur when combining a limber bottomhole assembly (BHA) with a PDC bit lacking some form of anti-whirl stabilization. On the basis of experience to date, these vibration-induced failures occur only in larger hole sizes. This paper describes the nature of these bit failures, their dependence on operating conditions, and practical recommendations for avoiding failure. Experience from one operation, in which revised drilling practices and bit selection criteria eradicated the failures, will be presented. Introduction The nature of various vibration modes, their damaging effects on downhole equipment, and the measurement and suppression technology developed to combat vibration, is well-documented in the literature.1,2,3,4,5 In drilling operations, one of the more common modes is torsional vibration, which is recognized by surface torque fluctuations with a frequency close to the fundamental torsional mode of the drillstring. The frequency of torque oscillations under such conditions is given by: f=12πkJb(1) Fig. 1, which was taken from an actual drilling operation, is an example of this frequency. These data, and all other vibration data described here, were acquired with a previously described detection system.5 In its extreme, torsional vibration can advance to the point that the bit and bottomhole assembly come to a complete stop-start (stick-slip) motion.6 Various characteristics in surface torque behavior can detect this extreme condition. First, the frequency of surface torque oscillations will decrease below the fundamental frequency, thereby reflecting the sticking time downhole. Second, peak surface torque will increase to a value approximately equal to the static torque required to break the bit/BHA free, plus the torque required to rotate the drillstring. In fact, if a drop in surface rotary speed accompanies this rise in torque, which is typical for topside rig equipment, the speed minima concurring with the torque maxima creates an inertial force opposite to string rotation. This adds another component to peak surface torque. Finally, the effect of inertia will also reduce surface torque below the level that is required to rotate the drillstring as the bit and BHA break free downhole. Fig. 2 is an example of this fully developed ‘slip-stick’ behavior. Of particular significance is the frequency of the torque oscillations compared to the calculated fundamental frequency. The rise in torque is linear to the observed maxima, reflecting the deceleration of the BHA during the stick phase. The subsequent torque drop suggests a rapid acceleration of the BHA during the slip phase. The purpose of this paper is to report the operator's experience with this form of slip-stick vibration in one major drilling program. The circumstances under which it occured, its effect on drilling performance/economics, and measures that have been developed to control it, will be described. In addition, the authors will use detailed analyses of surface drilling parameter and other operational data, and simulations using a finite element model of the drillstring to interpret the downhole behavior that provokes slip-stick.

Improvements in drilling efficiency and economics are continuously being achieved with PDC bits, specifically in medium-hard and non-abrasive formations - normal drilling. These gains have been aided by advancements in PDC bit technology, their selection practices and BHA design. However, PDC bit performance in harsh environments - hard, abrasive and heterogeneous formations - still lacks consistency and predictability. The problems associated with these applications are amplified when large hole sizes (diameters greater than 12") are drilled. Performance qualifiers such as footage, and especially rate of penetration (ROP), tend to be considerably lower in such instances. Considering the depths at which small hole sizes (diameters less than 9") are usually drilled, the likelihood of such hole sizes being drilled in harsh environments tends to be very high. It is evident that huge operational savings will be achieved if the performances of PDC bits are improved. This paper will discuss the effects formation hardness, abrasiveness and heterogeneity have on PDC bit performance. In addition, it will establish the influences different hole sizes have on PDC bits, especially in harsh environments. Drilling efficiency will be discussed in terms of operational parameters, vibrational behavior, durability equivalency (DEQ) and lithology differences. An engineered approach to PDC bit development and selection, that enhances performance in harsh environments and in large hole sizes, will be discussed. Background Researchers have identified stabilization1,2,3,4, durability5, directional efficiency6,7,8 and rate of penetration (ROP) as critical performance qualifiers (PQ) for PDC bits. Although the contributions of these qualifiers to operational success have normally been discussed in isolation, it is known that they exhibit strong dependencies. The importance of the different PQ's to PDC bit efficiency depends on the mechanical properties of the formations being drilled. Formation drillability9,10 (FD), characterized through drilling difficulty establishes the effects rock mechanical properties such as bulk compressibility, abrasivity, compressive strength and lithological composition have on the drilling process. Environments become harsh as their drilling difficulties increase. In some instances, high pore pressures, and associated higher mud weights, depositional depths and hole size compound the problems normally associated with harsh environments. These different PQs assume diverse roles when it comes to normal and harsh environment drilling. Most importantly, ROP's influence on operational efficiency in normal and harsh environments can be classified as direct and indirect respectively. This distinction establishes the groundwork needed to improve PDC bit performance in the challenging environments that will be described in this paper. Introduction Inaccurate evaluations of PDC bit dull grades have contributed to the known inefficiencies of these bits in harsh environments. In addition, inefficient interpretations of drilling parameters11,12, such as weight on bit (WOB), rotary speed (RPM), flow rate (Q) and HSI have also contributed to the performance inconsistencies of these bits.

M.J. Fear, SPE, BP Exploration Company (Colombia) Ltd. Abstract A method has been developed that identifies which factors are controlling rate of penetration (ROP) in a particular group of bit runs. The method uses foot-based mud logging data, geological information, and drill bit characteristics, to produce numerical correlations between ROP and applied drilling parameters or other attributes of drilling conditions. These correlations are then used to generate recommendations for maximising ROP in drilling operations. Introduction The time devoted to drilling ahead is usually a significant component of total well cost. On wells drilled by BP Exploration (BPX) in the last 5 years, this "rotating time" has accounted for 10-30% of the dryhole cost of the well. Because this rotating time is directly and inversely proportional to the ROP achieved by the drill bit, bit penetration rate has considerable significance in both control of drilling costs, and scope for cost reduction. Despite this, the interaction between the drill bit and rock, and the ROP that results; remains a part of the drilling process that is not particularly well understood in any level of detail. Other drilling phenomena, such as torque/drag behaviour, or the expected directional response of a bottomhole assembly, are typically managed with the aid of quantitative models, which have been validated against real well data and now offer a substantial improvement in understanding and reduction in uncertainty. Such tools are conspicuously absent from planning or analysis of bit performance. In the operational phase of drilling, similar uncertainty over expected ROP clouds decisions on which bit types to select, whether downhole motors or turbines could improve ROP, and which drilling parameters or bit nozzle arrangements would be most effective in maximising ROP. The contribution that intuition typically makes to these decisions testifies to the inability of drilling data or predictive models to provide anything more substantial. Against this background of commercial incentive and technical difficulty, BPX is involved in the development of methods which can be used to raise drill bit ROP. This paper describes one such method, and its application and benefits to a drilling operation. Background to the method Table 1 lists a number of factors which have been either proposed or observed to affect ROP. The number of factors is substantial, reflecting the complexity of the bit-rock interaction. This is compounded by observations that show significant inter-dependence between some of these variables, and non-linearity in some effects. Laboratory studies and modelling have however cast some light on how this complexity originates. First, how ROP responds to changes in drilling variables has been shown to depend strongly on the properties of the rock being drilled. In permeable rocks for example, overbalance pressure influences ROP, while this gives way to a dependance on bottomhole pressure as permeability decrease. Even overbalance pressure itself appears subject to dynamic influences as the bit advances, either via filtration effects on pore pressure in the zone being cut by the bit teeth, or stress effects on pore pressure around the wellbore. Bit cleaning problems while drilling hydratable formations in water based drilling fluids may also override the normal benefits of mechanical drilling parameters to ROP, so that rock mineralogy and drilling fluid chemistry are obviously significant "environmental" factors. These cleaning effects are though influenced by bit design and jet nozzle arrangement. In all, rock properties that influence bit response appear to include at least mineralogy, strength, density, porosity, and permeability. Given that these properties are not generally quantified while drilling in the field, it is not surprising that variations in ROP may be difficult to understand. Second, the effects of mechanical and hydraulic variables may also be intimately inter-dependant, so that the response of ROP to each of higher weight on bit (WOB), rotary speed and flow rate may vary with levels of the other parameters.