S

Figure 10-13. Effect of field mud filtrates on oil permeability of Paloma Field (Stevens sand) cores. (From Novak and Krueger.^ Courtesy API.)

- CONNATE WATER (IMMOBILE)

: IMMOBILE WATER-WET FINES

Figure 10-14. After oil has replaced aqueous filtrate, remaining fines are immobilized in connate water- (From MueckeJ7 Copyright 1978 by SPE-AIME.)

- CONNATE WATER (IMMOBILE)

: IMMOBILE WATER-WET FINES

Figure 10-14. After oil has replaced aqueous filtrate, remaining fines are immobilized in connate water- (From MueckeJ7 Copyright 1978 by SPE-AIME.)

reservoir, and the rate of production. Field studies have shown that productivity decreases at high rates of production,24 apparently because the high rate increases the concentration of tines, which favors bridge formation. In a micro-model study Muecke17 observed that once the oil-water interface had passed out of the model, no more fines were discharged. However, if oil and water were flowed simultaneously through the model, the fines continued to migrate indefinitely, because multiphase flow caused loca! pressure disturbances.

Permeability Impairment by Particles from the Drilling Mud

It is now well established that particles from the mud can invade the formation and cause impairment by blocking constrictions in the flow-channels. However, as we showed in Chapter 6, mud particles can only penetrate the formation during the mud spurt period, before the filter cake is established. Once the filter cake is fully formed, it filters out the finest colloids because of its structure and very low permeability (around 10md). The permeability may continue to decrease, but the decrease will be caused not by particles passing through the cake, but by transport and re-arrangenient of particles already carried in by the mud spurt.

It follows that the way to control mud particle damage is to minimize the mud spurt by ensuring that enough bridging particles of the right size are present in the mud. Bridging particles, it will be recalled from Chapter 6, fit into and block the surface pores of the rock, thus forming a base on which the filter cake can form. To be effective, the primary bridging particles must be not greater than the size of the pore openings and not less than 1 /3 that size, and there must be a range of successively smaller particles down to the size of the largest particles in the colloidal fraction. The greater the amount of bridging particles, and the lower the permeability of the rock, the quicker the particles will bridge, and the smaller will be the mud spurt -v

Since the mud spurt occurs when the formation is first exposed by the bit, tests for mud particle penetration should be made under the condition that the external cake is being continuously removed. Even under that condition, mud particles penetrate a remarkably short distance into a rock when adequate bridging particles are present. For example, tests by Glenn and Slusser10 showed mud particles penetrated 2 to 3 cm into alundum cores. Microbit tests by Young and Gray2 7 indicated that mud particles penetrated about 1 cm into Berea sandstone cores having a maximum permeability of 105 md. Krueger: N cites studies indicating particle penetration of 2 to 5 cm.

Most of the impairment caused by particle invasion is concentrated in the first few millimeters of the rock. For example, Young and Gray27 found the permeability of the first centimeter of the Berea sandstone cores to be reduced to about 10 2 md, and the remainder of the core to be essentially undamaged. In more permeable rocks, the permeability of the invaded zone beyond the first centimeter may be reduced to 70-80% of the original permeability.

Impairment that extends only a few centimeters into the formation can be eliminated by gun perforating. Klotz;ii et al have shown that formation damage can thus be eliminated provided that the length of the tunnel— normally about 8 in. (20 cm)—exceeds the depth of the damaged zone by at least 50% (see Figure 10-15). Therefore, we may conclude that damage from mud particles is not a cause for concern, provided that the mud contains adequate bridging particles and that the well is gun perforated (or under-reamed) with a non-damaging completion fluid which will be described later.

On the other hand, deep and irreversible damage will be done if adequate bridging particles are not present. An experiment by Abrams30 with a brine that contained no particles large enough to bridge illustrates an extreme case. The brine contained 1% of particles less than 12 microns, and was injected continuously into a 5 darcy sand pack in a radial system. Figure 10 16 shows the resulting severe impairment plotted as reduction in permeability versus depth of penetration. Backflushing with oil did little to restore permeability, and Abrams calculated that a similar impairment in a well with a drainage radius of 500 feet would reduce productivity to 14% of potential, compared to 99% which has obtained with a fluid containing bridging particles.

Krueger30a describes an experiment which showed very clearly that in order to avoid permeability impairment, either a properly filtered brine or a mud containing adequate bridging solids must be used; as shown in Figure 10-17, dirty brines cause severe impairment.

Figure 10-15. Effect of depth of gun perforations on well productivity. Note mud damage is virtually eliminated when the depth of the tunnel exceeds the depth of the invaded zone by 50%. K, is the permeability of the invaded zone as a percent of initial permeability. Percent well flow efficiency is ratio of the productivity of a damaged well to that of an undamaged well expressed as a percent, and calculated for radial flow. (From Klotz, Krueger, and Pye,29 Copyright 1974 by SPE-AIME.)

Figure 10-15. Effect of depth of gun perforations on well productivity. Note mud damage is virtually eliminated when the depth of the tunnel exceeds the depth of the invaded zone by 50%. K, is the permeability of the invaded zone as a percent of initial permeability. Percent well flow efficiency is ratio of the productivity of a damaged well to that of an undamaged well expressed as a percent, and calculated for radial flow. (From Klotz, Krueger, and Pye,29 Copyright 1974 by SPE-AIME.)

Distance into Model, inch

Figure 10-16. Productivity impairment by particle invasion when no bridging particles were present. (From Abrams.30 Copyright 1977 by SPE-AIME.)

Distance into Model, inch

Figure 10-16. Productivity impairment by particle invasion when no bridging particles were present. (From Abrams.30 Copyright 1977 by SPE-AIME.)

Occurrence of Mud Particle Damage in the Field

Any mud that has drilled more than a few feet will contain more than 1 lb/bbl {3 kg/m3) of particles in the size range of 50 to 2 microns, which is all that is required to bridge consolidated rocks of permeability less than about 1 darcy. In such formations, therefore, no special precautions are needed, and the productive interval can usually be drilled with the same mud that was used to drill the upper part of the hole. There are, however, a number of formations and types of operation in which adequate bridging requirements are difficult to estimate, and sizes larger than 50 microns may be required. Under the circumstances discussed below, steps must be taken to assure adequate bridging; when that cannot be assured, it is advisable to use a special completion or workover fluid whose solids degrade or may be dissolved when the operation is completed.

Unconsolidated Sands. Unconsolidated sands often require particles larger than 50 microns to bridge. Because of the wide range of particle and pore sizes and shapes, it is difficult to specify sizes and amounts for bridging, but 5 to 10 lb/bbl (15 30 kg/m3) of bridging particles with a maximum size of 150

0 0

Post a comment