Reservoir Completion 69, Perforation skin Shot density (spf) Figure 2.53. Karakas and Tarig perforation skin ~ sensitivity to phasing angle slightly improve productivity. As Figure 2.53 shows there is little difference between 45°, 60°, 90° and 120° phasing angles. Determining the input parameters for the model depends on good shoot test data such as API RP19B section 4 tests, with any corrections for variations in rock strength, pressure, standoff, etc, Section 1 data should never be used directly. For other parameters, Pucknell and Behrmann (1991) suggest that the crushed zone thickness is around 0.25-1 in. with the greater thickness for larger charges (22 g charges). They reported a large variation in crushed zone permeability with permeability reductions in the range of 50% to 80%. API 19B section 4 perforation tests also allow some determination of the effect of the crushed zone on productivity, Some degree of caution is required when using models such as these to accurately predict skin. As we have scen, predicting many of the parameters such as crushed zone, damage and especially perforation length is problematic, In a real- world situation, many perforations are also plugged by debris. The real benefit of the models is in making comparisons between options demonstrating that perforation length is critical in getting low skins. The models, for example, can thus be used to compare the benefit of additional charge weight versus a corresponding reduction in shot density. Other, more recent, models are available. For example, Hagoort (2007) presents a model that better models the flow into the perforation tip and includes non-Darcy effects 2.3.5. Perforating interval selection It is usually the job of the reservoir engineer to select the correct perfo- rating interval. However, some assistance from completion/petzoleum engineers70 Perforating is beneficial. 1, How accurate is the depth correlation — how accurate does it need to be? 2. Are sufficient intervals of good quality cement bond left for possible future isolation opportunities such as setting bridge plugs? ‘What is the optimum order for multiple interval perforating? . Are the needs of subsequent stimulation being addressed? For wireline-deployed perforating, depth correlation is usually with the aid of gamma ray and casing collar locator (GR/CCL). These are tied into the open hole logs. Depth contol is achievable to an accuracy of less than 1ft. For tubing- conveyed perforating, depth control can be achieved by dead reckoning, but stretch, thermal expansion and drag, not to mention human error, will limit the accuracy to tens of feet at best. It is common to use an electricline correlation run through the tubing prior to picking up the tubing hanger or test tree. The correlation run can tie into the open hole logs above the reservoir or a radioactive pip tag strategically positioned. For a deepwater completion, the water depth will limit the uscfulness of the correlation ran, For a completion with a hydraulic set packer or dynamic seals, tubing movement should be accounted for, although the packer movement during setting is typically only a few feet (Section 9.12.1, Chapter 9). For coiled tubing and sickline-deployed guns, it is possible to improve the accuracy by using a memory GR/CCL run. Memory effect on the coil along with changes in geometry or weight can interfere with this method. Without a memory run, accuracies of tens of fect are still possible with slickline typically being accurate to around +1 per 1000 ft (King et al., 2003). There are also a number of devices that can assist with both coiled tubing and slickline perforating accuracy 1. Tubing end locator — a lever that latches the end of the tubing and thus creates an overpull when the assembly is pulled back. 2. A depth correlation sub ~ a profile that matches the geometry of a roller ~ again a small overpull is noticed when the assembly drops into this profile. 3. Tagging the bottom of the well — simple but not without the risk of getting stuck, 4, Slickline collar locators, ‘These sophisticated devices (at least for slickline) use a standard electronic Ci ~ this works by sensing changes in the magnetic field. The signal is processed and then converted to additional tension in the string by a drag mechanism (Foster et al., 2001). Such a system will also detect components such as nipples or other completion equipment. 5. Coiled tubing pulse telemetry. Coiled tubing has the advantage of incorporating a flow path that can be used in the same way that measurement and logging while drilling (MWD and LWD) tools can transmit data to surface during drilling operations. Logging information such as GR or CCL can be converted to digital data and transmitted to surface by temporarily restricting the flow going through the bottom hole assembly. The resultant pressure pulse is picked up at surface. These systems can be run in conjunction with conventional drop ball hydraulic firing heads without interference (Flowers and Nessim, 2002)Reservoir Completion n For clectricline capability in a high-angle well, coiled cubing can be prefitted with electricline (so-called stiff electricline) although the use of tractors has reduced this application. Iv is also worth asking how accurate the perforating needs to be. Although it makes no sense to perforate shales and other non-productive intervals, providing 2 generous overlap will require much lower accuracy than trying to precisely perforate each small interval. The exception is where water or gas has to be avoided or for the later setting of bridge plugs in unperforated intervals, ‘Where future setting of bridge plugs is required, it is usually accepted that an unperforated interval with a quality cement bond of around 10-15 f is required. Assessing the cement bond is a notoriously difficult subject, with well-known issues such as micro annuli interfering with the interpretation. A detailed consideration is beyond the scope of this book, but from personal experience apparently ‘free’ pipe can casily be reinterpreted to give a quality cement job! 2.3.5.1. Perforating for stimulation ‘The topic of the optimum perforation design for fracture-stimulated wells is often discussed, with opinions divided. It is proven that poor perforation design can lead to poor stimulation — particularly with the bigger treatments and reduced polymer loadings that are now common. In particular, poor perforations increase the risk of screen-out through increased tortuosity, back-pressure and the generation of multiple fractures. Itis also clear that the best practice for perforating for stimulation is not the same as perforating for non-stimulated wells. Given that a fracture usually has a strongly preferred propagation direction, it would make sense to orientate the guns along the preferred propagation direction. Behrmann and Nolte (1999) suggest that a tolerance of +30° is required. Methods for achieving this are discussed in the section on oriented perforating for sand control (Section 3.2.1, Chapter 3). Unfortunately, the stress directions are not always known precisely or the well is deviated. Perforating at 60° or 45° phasing covers most cases — at the expense of most of the perforations not accepting proppant or being in poor communication with the fracture, This places more importance on adequate (open) perforations. However, perforations that are fully open and completely free of debris are not required — dynamic underbalance perforations may be perfectly adequate. Extreme overbalance or propellant perforating may also be effective in clearing the entrance hole of perforating debris. Given that the fracture will usually initiate at the cement—rock interface, an adequate entrance hole is required to prevent bridging — but this does not necessarily mean big hole charges. Big hole charges create a greater stress cage around the perforations and therefore a more tortuous path fom the perforation to the fracture Pongratz ct al., 2007). There is also a risk with big hole charges under downhole conditions that the perforation tunnel length will be inadequate — especially where the guns are not centralised. This is again particularly the case for the deep, low-permeability, high-strength reservoirs that arc so often the target for stimulation. Fracturing for sand control (fac packs) is a completely different