1st National Iranian Drilling Industry Congress.

Bit Run Optimization through Simulation

Mohsen Ebrahimi Young researchers club, Islamic Azad University, omidieh Branch Isa Noveiri Production technology research institue

Abstract: Weight on bit, rotary speed, bit hydraulics and more importantly bit types, are the most important parameters affecting rate of penetration and consequently the economics of drilling. Pulling a worn bit is necessary to make the drilling operations more efficient. Cost per foot calculation is a popular criterion used to predict when to pull a bit. In this paper, some first 3 thousands feet of a reservoir lithology including four bit runs are analyzed so that factors affecting penetration rate are optimized and the lowest possible well cost is achieved. The effects of using other bit types, weight on bit, rotary speed and bit hydraulics are simulated in an attempt to determine the optimized parameters, which result in the lowest drilling cost. Overall, it was found that in bit runs 2, 3 and 4 some lower cost per foot can be achieved using milled tooth bits instead of tungsten carbide types. Furthermore, it can be concluded that by increasing weight on bit and reducing rotary speed in harder rocks, more favorable results are obtained while in softer rocks it is vice versa. In addition, by lowering both mud flow rate and nozzle sizes, hydraulic horse power dissipated at the bit is increased, hence decreasing cost per foot. Keywords: Bit Run Optimization, Simulation, Weight on bit, Rotary Speed, Bit hydraulics

www.petroman.ir

1st National Iranian Drilling Industry Congress.

1. Introduction: The usual objective when drilling a well is to drill for the lowest overall cost. At first sight, this may seem like a requirement to drill as fast as possible, since many of the costs, such as the rig day rate, wages etc., are time dependent. However, some costs are fixed, for example the cost of the bit, and it may not always be economic to pay a very high cost for a bit that drills somewhat faster, particularly if the other costs are low [1]. Many parameters affect the performance of a drilling operation, that if not adjusted properly, make the operation uneconomical. Weight on bit (WOB), rotary speed (RPM), bit hydraulics and more importantly the type of the bits used, are the most important parameters affecting rate of penetration and consequently the economics of drilling. However, all the optimization in the world will not make a bit last for the entire well [2]. The most common criterion to determine the best time to terminate a bit run is the cost per foot calculation. The drilling simulator calculates the cost per foot for each bit run, re-setting the calculation for each new bit. It takes into account the time taken to change the bit, and calculates the time taken to trip into and out of hole. The other costs include the bit cost and the rig day rate (including all the associated overheads) [1]. In this paper, some first 3 thousands feet of a reservoir lithology including four bit runs and a 13 3/8" casing set at 2600 feet is analyzed so that the factors affecting rate of penetration are optimized and the lowest possible well cost is achieved. All bits used in the original drillings process were of tungsten carbide insert type. In the first step, other drilling bit types were examined to see if a better drilling performance is observed. Then, the effect of altering operating conditions (WOB and rotary speed) has been investigated. At the end, it was attempted to get even better results through bit hydraulics optimization. As a result, it was found that runs number 2, 3 and 4 could yield a lower cost per foot using milled tooth bits because they have higher rate of penetration and also cost less than half the price of a tungsten carbide bit. In addition, it is shown that altering the weight on bit is more effective than a change in rotary speed. Optimized drilling in harder sections of lithology demands a higher weight on bit while the same is not true for softer sections. Apart from that, lowering both mud flow rate and nozzle sizes increases hydraulic horse power dissipated at the bit, hence decreasing cost per foot. Optimization of these 4 factors will save us about 17% of the original well cost. 2. Factors Affecting the Rate of Penetration (ROP): The factors that influence the rate of penetration (ROP) during a conventional rotary drilling process are numerous. In general these factors can be classified into six groups [3, 4]: 1. Bit type 2. Formation characteristics 3. Drilling fluid properties 4. Bit operating conditions (weight on bit and rotary speed) 5. Bit tooth wear 6. Bit hydraulics

www.petroman.ir

1st National Iranian Drilling Industry Congress.

2.1 Bit Type: The bit type selected has a large effect on penetration rate [3]. PDC bit drills the fastest in the soft shale, and also penetrates rapidly in the soft sand, but it has difficulties in penetrating the harder sand and the limestone. The milled tooth bit does well in the softer rocks also. In the harder rocks, it does relatively less well. The milled tooth bit does not do quite as well as the PDC bit in the shale, where it tends to suffer from the accumulation of sticky clay cuttings between the bit teeth, a problem known as bit balling. The tungsten carbide insert bit has shorter teeth than the milled tooth bit, and it is thus not aggressive enough to drill rapidly in the softer formations. It also suffers from bit balling in the shale section. However, its great advantage lies not so much in a high rate of penetration as in its greater abrasion resistance, and therefore longer life, particularly in the harder rocks. The natural diamond bit has an overall lower rate of penetration than any of the other bits. As with the tungsten carbide bit, its great advantage lies in its ability to penetrate the very hardest rocks with acceptable wear [1]. Unfortunately, the selection of the best available bit for the job can be determined only by trial and error. The most valid criterion for comparing the performance of various bits is the drilling cost per unit interval drilled [3]. The results of the cost analysis sometimes must be tempered with the engineering judgment. Reducing the cost of a bit run will not necessarily result in lower well cost if the risk of encountering drilling problems such as stuck pipe, hole deviation, hole washout, etc., is increased greatly [3]. 2.2 Formation Characteristics: Elastic limit and ultimate strength of the formation are the most important formation properties affecting penetration rate. Permeability of the formation and mineral composition of the rock can also has major effects on penetration rate [3]. However, these factors are beyond our control and can not be manipulated; therefore no optimization can be made on them. 2.3 Drilling Fluid Properties: ROP tend to decrease with increasing fluid density, viscosity and solid content, and tend to increase with increasing filtration rate [4]. These parameters usually depend on the nature of pore and fracture pressures of the formation and hence no optimization is performed on them. 2.4 Bit Operating Conditions: WOB and rotational speed (RPM) of the drill string have a major effect on both the penetration rate and the life of the bit. In addition, these parameters can be varied easily. Thus, the determination of the best weight and rotary speed for a given bit is one of the routine problems faced by the drilling engineer. Generally speaking, no significant penetration rate is obtained until a threshold weight on bit is applied. Penetration rate then increase rapidly with increasing values of weight on bit. However, at higher values of weight on bit subsequent increase in it causes only slight improvements in rate of penetration. In some cases a decrease in

www.petroman.ir

1st National Iranian Drilling Industry Congress.

rate of penetration is observed at extremely high values of weight on bit, Attributed to less efficient bottom hole cleaning [3]. Penetration rate usually increases linearly with rotary speed at low values of rotary speed. At higher values of rotary speed, the response of penetration rate to increasing rotary speed diminishes. The poor response of penetration rate at high values of rotary speed usually is also attributed to less efficient bottom hole cleaning [3]. 2.5 Bit Tooth Wear: Most bits tend to drill slower as the bit run progresses because of tooth wear. Bit wear is often directly dependent on WOB, rotary speed and the rock type being drilled. Operating conditions and bit wear are simultaneously optimized through using cost per foot calculation. 2.6 Bit Hydraulics: The efficiency of cleaning the bottom hole does not just depend on the mud flow rate, but also on the degree of local turbulence at the bottom hole. However, if the power and maximum pressure of the mud pumps are limited, it is intuitively clear that there must be some optimum combination of flow rate and pressure drop across the nozzles to get the best cleaning effect [1]. There is still discussion as to which combination of pressure and flow rate produces the best cleaning effect. The most popular criteria assume that the cleaning effect is maximized either when the hydraulic horsepower dissipated at the bit is a maximum, or when the change in momentum of the mud as it hits the hole bottom is a maximum. The latter is often known as the "Jet impact" criterion [1]. The hydraulic horsepower dissipated at the bit nozzles is: Eq (1) Bit Hydraulics Horse Power = C1 Pbit Q Where C1 is a constant, Pbit is the pressure drop over the nozzles and Q is the mud flow rate. For practical purposes, however, there does not appear to be a great difference between the two criteria, and so the hydraulic horsepower criterion has been chosen for the simulator as it is more easily comparable with the mud pump horsepower [1]. However, having simulated different scenarios, it was found that this criterion does not necessarily result in the lowest cost. Therefore, the cost per foot calculation is considered as the basis of selecting the best combination. 3. Cost per Foot Analysis: There is almost always some uncertainty about the best time to terminate a bit run and begin tripping operations. If the lithology is somewhat uniform, cost per foot calculation can be used as a criterion. In this case the best time to terminate the bit run is when the lowest cost per foot is achieved. However, when the lithology is not uniform, this procedure will not always results in the minimum total well cost. In this case, an effective criterion for determining optimum bit run is obtained only after enough wells are drilled in the area to define the lithologic variations. For example it is sometimes desirable to drill an abrasive

www.petroman.ir

1st National Iranian Drilling Industry Congress.

formation with an already dull bit and then place a sharp bit in the next shale section. Alternatively, it may be best to terminate a bit run in order to place a hard formation bit in an extremely hard abrasive section [3]. Costs are usually broke into two categories: (1) Fixed costs and (2) Variable operation costs [3]. The usual objective when drilling a well is to drill for the lowest overall cost. At first sight, this may seem like a requirement to drill as fast as possible, since many of the costs, such as the rig day rate, wages, etc., are time dependent. However, some costs are fixed, for example the cost of the bit, and it may not always be economic to pay a very high cost for a bit that drills somewhat faster, particularly if the other costs are low [1]. Unless the bit run is to be terminated for a specific reason, such as logging or casing the well, the cost of each bit run can be minimized by calculating the cost per foot as the hole gets deeper. This is done by summing the fixed and time-dependent costs and dividing by the total footage drilled during the bit run [4]:
Cost Bit cost + (Drilling time + Trip time) Rig cost p er hour = ft Feet drill ed this bi t run

Eq (2)

When we run in hole with the new bit, we have already incurred the cost of the new bit and spent time to run in hole, but have drilled no distance. Our cost per foot is therefore infinite. However, as soon as we begin drilling, the factor "Feet drilled this bit run" begins to increase, and so the cost per foot decreases. As we continue drilling, the fixed costs remain constant, but the time related costs increase, as does the footage drilled. Initially, the fixed costs are greater than the time-related costs, but eventually the time-related costs begin to dominate the top line of the expression. In the real situation, the bit gradually wears, and so the number of additional feet drilled per additional hour gets less as time increases. The timerelated costs continue to increase steadily, however, and so eventually the cost per foot reaches a minimum and then begins to rise again. It is at this minimum value that we have reached the minimum cost per foot for the bit run, and we should therefore replace the bit. Finally, note that the Cost per Foot calculation can be used even if the bit run has to be terminated for other reasons. In this case, all that needs to be done is to find the set of conditions that give the minimum cost per foot at the depth where the bit run is to be terminated [1]. 4. Drilling Simulator: The simulator (called "Payzone" for short), is a computer program that receives: a description of a series of rock layers (lithology), a description of one or more drill bits, and a set of operating parameters such as weight on bit, bit rotary speed, mud flow rate and other required information, as input. The simulator then calculates the rate of penetration and the rate of wear of the bit. From this information, a plot of drilled depth versus time is obtained [5]. Two parameters govern the drilling response. These are the "Softness Factor" (S) that determines the initial rate of penetration and the "Wear Factor" (W) that determines the rate of bit wear. As drilling proceeds, the drilling model is

www.petroman.ir

1st National Iranian Drilling Industry Congress.

continuously calculating the state of wear of the bit and its rate of penetration. The wear algorithm starts by calculating a "wear factor". At each time step the "Life" of the bit is multiplied by the "wear factor" to reduce the life by a certain fraction of its current value as the time increases. To obtain the current rate of penetration, the present "Life" is combined with two more factors that present that represent the interaction of the particular bit with the rock being drilled, and the underlying rate of penetration. The result is further combined with the factors that depend on the weight on bit, the rotary speed and the mud density. These operations result in the calculation of rate of penetration that would be characteristic of a bit for which the cuttings are removed as soon as they are formed (Perfect Cleaning). At this point the effect of mud flow rate is introduced, and a final rate of penetration is calculated based on the effectiveness of the mud flow in cleaning the cuttings away from the bottom hole [5]. The simulator has been given a capability to infer rock strength from sonic data following recommendations available in the literature, if strength data are not available; however, to estimate abrasivity, it is necessary to define a way to estimate this value. An example of how the abrasivity of the formation is estimated is shown in Eq. 3 [6]:

Abrasivity = Constant Quartz Con tent Rock Stren gh

Eq(3)

5. Case Study: 5.1 Lithology: For current study, the first 3465 feet of a reservoir lithology including four bit runs and a 13 3/8" casing set at 2600 feet are analyzed so that the factors affecting the rate of penetration are optimized and the lowest possible well cost is achieved. Table 1 shows the lithology of the reservoir where each layer is characterized by its number, depth, thickness, fracture gradient, its drilling softness (S) and abrasivity (W), the three logging values of Gamma Ray Activity (GA), Resistivity (Res), Porosity (Por), the fluid type and the pore fluid pressure gradient.
Table 1: Lithology of the section being studied

www.petroman.ir

1st National Iranian Drilling Industry Congress.

5.2 Bit Run Optimization Methodology : All the bits used in the original bit runs were of tungsten carbide insert type. The first 2600 ft of the lithology was drilled using a 17 1/2" bit and mud density of 9 ppg. At the depth of 2600 feet, a 13 3/8" casing was set, mud density was increased to 10.2 ppg and 12 1/4" bit was used. Table 2 shows the 4 bit runs coupled with their resultant cost per foot and overall well cost. As it can be seen, drilling this section takes 163 hours and imposes a cost of about 325000 dollars.
Table2: Original Bit runs and their associated costs
Bit Bit Bit Run Type Size # 1 2 -3 4 T.C 17 1/2" Depth In (ft) 0 Depth Out (ft) 1325 2600 Time Mud Nozzles Mud Rotary WOB Taken Weight Size Flowrate Speed (lbs) (hr:min) (ppg) (32nds) (gpm) (RPM) 38:12 ' 70:17 ' 9 9 -10.2 10.2 15 15 -15 15 760-750 750-730 -675 675 100 85 -80 120 38000 40000 -40000 47000 Overal Cost/ Foot 57 55 94 87 94 Well Cost ($) 75699 143805 243828 257919 325532 Saving ($)

------

T.C 17 1/2" 1325

Casing 13 3/8" set at 2600 ft 107:16 ' T.C 12 1/4" 2600 T.C 12 1/4" 2965 2965 3465 118:30 ' 163:12 '

In the first step, it was attempted to obtain a better cost per foot results with other bit types, keeping operating parameters as the original bit runs. To avoid complexity, only 4 main bits types with predefined properties were analyzed. Table 3 shows the ROP and Wear Factor of the bits studied. The total wear factor is the result of first combining this wear factor and wear parameter entered in the lithology, and then further combining with two parameters that represent effects of weight on bit and rotary speed on the rate of wear of the bit [5]. Having determined the best bit types to be used, different sets of operating parameters (WOB, rotary speed and bit hydraulics) were simulated to see if a better cost and time saving can be yield. The procedure consists of a step-by-step approach, simulating different scenarios in sequence.
Table 3: ROP and Wear Factor of different bit types
Bit \ Rock PDC M.T T.C Diamond ROP Factor Wear Factor ROP Factor Wear Factor ROP Factor Wear Factor ROP Factor Wear Factor Shale 20 20 11 280 8 25 3 5 L'Stone 14 210 13 220 12 70 4 2 S.Sand 17 200 20 550 15 100 4 100 H.Sand 17 400 12 350 12 90 7 55 Chert 5 250 6 300 6 50 5 60

www.petroman.ir

1st National Iranian Drilling Industry Congress.

Since there are many shale layers in the lithology, the well has been drilled using an oil based mud. The Mud pump is limited to provide 900 Horse power (HP). This value has been taken into account in the simulation.

5.3 Results and Discussion: The first step consisted of comparing other bits cost per foot results to find a more economic case. Results are shown in table 4 where, regarding bit types, only the first bit run was recognized optimized. Runs number 2, 3 and 4 could yield a lower cost per foot using milled tooth bits because they have higher rate of penetration and also cost less than half the price of a tungsten carbide bit. Since Tungsten carbide bits have more bit wear tolerance, they may seem more efficient to be used for drilling harder rocks. However, the distance to be drilled in the fourth section (hard lime stone) is not that long to justify higher cost of the bit.
Table 4: Optimization of bit types
Bit Bit Bit Run Type Size # 1 2 T.C 17 1/2" Depth In (ft) 0 Depth Out (ft) 1325 2600 Time Mud Nozzles Mud Rotary WOB Taken Weight Size Flowrate Speed (lbs) (hr:min) (ppg) (32nds) (gpm) (RPM) 38:12 ' 69:35 ' 9 9 15 15 760-750 750-730 100 85 38000 40000 Overal Cost/ Foot 57 48 87 81 85 Well Cost ($) 75699 Saving ($)

M.T 17 1/2" 1325

125924 17881 225947 17881 239809 18110 295105 30427

Casing 13 3/8" set at 2600 ft 106:34 ' 3 4 M.T 12 1/4" 2600 M.T 12 1/4" 2965 2965 3465 117:49 ' 158:35 '

-10.2 10.2

-15 15

-675 675

-80 120

-40000 47000

In the second step, an attempt was made to improve drilling performance through optimizing weight on bit and rotary speed. The results of optimization are summarized in table 5. Overall, sections 1 and 2 (bit runs 1 and 2) have rocks of different strength and abrasiveness; therefore, they can be interpreted as rocks with moderate strength and abrasiveness. In such formations, a moderate rotary speed with moderate WOB should maximize the drill rate and reduce bit wear. Bit run 3 drills a section of shale and soft sandstone. It can be inferred that in soft rocks, increasing the rotary speed and relatively low WOB can improve the penetration rate. Note that the weight on bit of 45000 pounds for 12 1/4" bit is considered lower than 42000 pounds for 17 1/2" bit. In harder formations just like section 4, increased WOB is more important than increased rotary speed. Moderately high WOB with low rotary speed yields the most optimum drill rates.

www.petroman.ir

1st National Iranian Drilling Industry Congress.

Table 5: Optimization of WOB and rotary speed
Bit Bit Bit Run Type Size # 1 2 T.C 17 1/2" Depth In (ft) 0 Depth Out (ft) 1325 2600 Time Mud Nozzles Mud Rotary WOB Taken Weight Size Flowrate Speed (lbs) (hr:min) (ppg) (32nds) (gpm) (RPM) 36:05 ' 64:18 ' 9 9 -10.2 10.2 15 15 -15 15 760-750 750-730 -675 675 100 120 -130 100 42000 42000 -45000 62000 Overal Cost/ Foot 55 46 84 78 81 Well Cost ($) 73056 Saving ($)

2643

M.T 17 1/2" 1325

119328 24477 219351 24477 231642 26277 280084 45448

Casing 13 3/8" set at 2600 ft 102:44 ' 3 4 M.T 12 1/4" 2600 M.T 12 1/4" 2965 2965 3465 111:05 ' 146:38 '

At the end, bit hydraulics was optimized. The data are shown in Table 6. The nozzle sizes 8/32 to 12/32 of the third and the fourth bit runs did not actually change cost per foot calculated and the biggest size is reported here.
Table 6: Optimization of Bit Hydraulics
Bit Bit Bit Run Type Size # 1 2 T.C 17 1/2" Depth In (ft) 0 Depth Out (ft) 1325 2600 Time Mud Nozzles Mud Rotary WOB Taken Weight Size Flowrate Speed (lbs) (hr:min) (ppg) (32nds) (gpm) (RPM) 34:27 ' 62:23 ' 9 9 -10.2 10.2 10 10 -12 12 456 453 -535 533 100 120 -130 100 42000 42000 -45000 62000 Overal Cost/ Foot 54 45 84 76 78 Well Cost ($) 71017 Saving ($)

4682

M.T 17 1/2" 1325

116935 26870 216958 26870 225340 32579 270270 55262

Casing 13 3/8" set at 2600 ft 100:01 ' 3 4 M.T 12 1/4" 2600 M.T 12 1/4" 2965 2965 3465 107:42 ' 141:24 '

Figures 1 and 2 show the overall well cost and savings made due to optimizations for the studied bit runs. Figure 3 compares original and optimized drilling operation in terms of time taken. Having made these optimizations, 55262 $ of the total well cost is saved which is about 17% of the original well cost.
400000 300000 200000 100000 0
Bit Run 1 Bit Run 3

Overall Well Co st ($)

Origin al Opt . (Type) Opt . (WOB&N)

Figure 1: Overall well cost for different bit run optimizations

www.petroman.ir

1st National Iranian Drilling Industry Congress.

Figure 2: The amount of money saved due to different bit run optimizations

Figure 3: Comparison of the original and optimized drilling

6. Conclusion: Simulation is a powerful tool to enhance drilling performance and thereby reducing total drilling cost. A simulator named payzone can fulfill this task using an mechanistic algorithm. Overall, in this study milled tooth bits are preferred to other bits, because it can yield greater ROP with an acceptable wear. The drilling parameters analysis showed that WOB and ROP are critical in drilling optimization. In soft sections of the lithology, increasing the rotary speed can improve the penetration rate with little effect on bit cutter wear. Relatively low WOB is recommended. In medium strength sections increased rotary speed will not have the same result as in soft formations. Moderate rotary speed with moderate WOB should maximize the drill rate and reduce bit wear. In harder portions, increased WOB is more important than increased rotary speed. Moderately high WOB with minimum rotary speed yields the most optimum drill rates. Due to mud pump limitations in providing high pressures, mud flow rate and nozzle sizes should be adjusted in a manner that the overall cutting removal is the nearest possible case to perfect cleaning process. In this study, by reducing both nozzle sizes and mud flow rate, the horse power dissipated at the bit is increased, hence reducing cost per foot of interval drilled. As it is noted, by using this simulator, 17% of the original well cost is saved, which is a considerable amount of money.

www.petroman.ir

1st National Iranian Drilling Industry Congress.

7. References: 1- Payzone user manual. 2- Baker Hughes INTEQ, Drilling Engineering Workbook, Baker Hughes INTEQ, (1995). 3- Bourgoyne, A., Millheim, K., Chenevert, M., Young, F.: Applied Drilling Engineering, society of petroleum engineers, (1986) 209. 4- Lyons, E.: Standard Handbook of Petroleum and Natural Gas Engineering, Volume 1, Gulf Publishing Company, (1996) 1090. 5- Cooper, G.A., Cooper, A.G. and Bihn, G.: An Interactive Simulator for Teaching and Research, paper SPE 30213 presented at the 1995 Petroleum Computer Conference, Houston, 11-14 June. 6- Abouzeid, A.A. and Cooper G.A.: The Use of a Drilling Simulator to Optimize a Well Drilling Plan, paper presented at the 2001 Geothermal Resources Council Annual Meeting, San Antonio, Texas, 26-29 August.

www.petroman.ir