REVERSE CIRCULATION AIR DRILLING CAN REDUCE WELL BORE DAMAGE

Reuben L. Graham, John M. Foster, Paul C. Amick, and J. Stanley Shaw
Reuben L. Graham Inc. Charleston, W.Va.

Reverse circulation air drilling coupled with an air dryer at the surface helped eliminate formation damage in several gas wells.

During reverse circulation drilling, the air flows down the annulus and up the drill pipe (Fig. 1).

The following were the three primary objectives of damage-free drilling (DFD):

Reducing damage so the initial open flows would more accurately reflect natural permeability
Reducing damage for diagnostic tools (temperature logs, noise logs, mud [gas composition] logs, and bore-hole television) to better detect liquid and gas entry points
Improving sampling by returning larger cuttings with shorter and more precise lag times.
The secondary objectives were to reduce drilling costs through the following:
Lowering required circulating air volumes when cuttings are reversed up the drillstring
Reducing water influx from shallow water zones because of annular back pressure during circulation
Improving penetration rates for larger holes.

Since 1985, Reuben L. Graham Inc. (RLG), sponsored by the Gas Research Institute (GRI), has researched the problem of identifying small shows during drilling. The elimination or reduction of formation damage during drilling would allow gas or oil shows to be identified better and quantified for subsequent completion plans.

Well bore damage or near well bore permeability reduction from normal mud drilling is easily understood by considering mud filtrate invasion and the resulting filter cake deposited on the well bore. The damaged condition of the well bore can be related to an extra pressure drop (or positive skin) from the drainage boundary to the well bore needed to maintain a given flow rate.

Well bore damage from rotary air drilling is less than that from rotary mud drilling or cable tool drilling, but the damage may still be enough to hinder the ability of a low pressure, low permeability zone to flow at its undamaged potential. Normal air drilling can result in a dust storm of cuttings moving at 35-40 mph in the drill pipe annulus and at 30-60 mph around the drill collars. This dust storm can cause damage to the formation, or "dusting-off." The drilled fines coat the well bore and possibly enter the pore throats. Additionally, "mudding off" of the well bore can occur when humid air is used as the circulating medium.

Believing mudding off to be the more damaging of the two, RLG developed an air dryer to eliminate free water in the compressor air stream and prevent mudding off of the borehole wall during air drilling.1 Although the research work determined that use of the air dryer resulted in a cleaner hole, it was difficult to establish any qualitative value for the reduced damage.2 Damage free drilling (DFD) by reverse circulation was thought to eliminate both dusting off and mudding off problems.

Before the DFD experiments began, several criteria were selected to compare and contrast DFD with normal drilling operations. The five validation tools (or tests) to evaluate the effectiveness of DFD were tests while drilling, analysis of borehole television pictures, open hole permeability tests, analysis of sample cuttings size and quality, and analysis of drilling parameters.

TESTS WHILE DRILLING

Probably the most useful and simplest tool to be used in evaluating formation responses to different air-drilling methods is the test while drilling (TWD).

The TWD is conducted by blowing the hole clean of cuttings and then stopping air circulation. Flow is prevented up the drill pipe by a bit float during normal circulation and a kelly cock during reverse circulation. The blowout preventer is closed around the drill pipe and any flow from the well bore is directed out the kill line. An orifice well tester and chart recorder measure and record flow rates. The well is permitted to flow until the rate stabilizes, usually after about 15-20 min.

When the drillstring is out of the hole, the blind rams are closed on the BOP, and the rates are measured through the kill line as described above.

TWDs helped investigate the incidence and decline characteristics of gas shows which would go unnoticed under normal rotary drilling operations. Three possible explanations for this phenomenon were considered:

Air volumes and high penetration rates mask the incidence of small volume gas entries
Some form of damage occurs at the well bore face, restricting gas entry
Hydrocarbon entries decline in rate to insignificant levels before the well reaches total depth. These entries were identified and continuously monitored by the mud log. Although zones with large open flows are easily detected in an empty air-filled hole, the TWDs dealt with identifying smaller, potentially contributing reservoirs which should be included in normal completion plans.

BOREHOLE TELEVISION

One of the more interesting and informative tools used in investigating air-drilled well bores is the borehole television (BHTV). This tool, which was improved through the GRI Devonian gas shales program and is now accepted and used throughout the industry, allows the borehole features to be viewed directly at the surface.

The BHTV is run by a wire line truck with special coaxial logging cable. The state-of-the-art camera provides excellent real-time images which are depth-encoded and oriented by an onboard gyroscopic compass. The logging run, which is usually accompanied by a correlation gamma ray log, is recorded on regular VHS video tapes for later review or computer enhancement.

In this particular application, the BHTV was used to locate gas, oil, and water entries and to describe the general condition of the well bore. The tool is capable of seeing small fractures, bedding planes, faults, and liquid entries that would be passed up by the normal logging program used in the Appalachian basin. The BHTV can determine the relative cleanliness of the borehole walls, both before and after DFD operations.

PERMEABILITY TESTS

Open hole permeability tests (OPT) were planned to determine the physical properties of small selected intervals. From these tests, values for permeability, skin, and reservoir pressure can be determined.

These OPTs were nitrogen injection/falloff tests conducted in small, 11-ft intervals isolated by inflatable packers run in the hole on drill pipe. These tests, an integral part of the GRI research, are a combination of a drill stem test and a nitrogen slug test. (A topical report to GRI on OPTs will be released in the near future.)

The results of the OPTs are important to DFD; the tests conducted on the three wells proved that a damaged condition of the well bore (as indicated by positive skin factors) existed in intervals drilled using normal drilling techniques.

CUTTINGS SAMPLES

The cuttings samples were taken from the blooie line. Cuttings size is an indicator of reverse circulation effectiveness. The larger fragments recovered at the surface illustrate that hole cleaning is handled adequately by the bit design.

The quality of the sample provides an indication of cuttings fall back and contamination. These two problems should be greatly reduced or eliminated with continuous circulation and shorter, precise lag time.

DRILLING PARAMETERS

Several rig operating parameters were watched closely during reverse circulation drilling because they provided indications of how well the experiment progressed in real time. The most obvious effect on drilling is penetration rate. During these tests, other operating parameters (bit weight and rotary speed) were kept close to normal drilling values for proper evaluation of the effects of reverse circulation. Therefore, higher penetration rates were attributed to better hole cleaning and lower rates to poor cleaning.

Any variation in circulating air pressure was also a focal point. High pressures indicated restrictions (hence, lower velocities) or plugging. Low pressures indicated higher air velocities and better cuttings carrying capacity.

Compressor rates were also monitored to determine the optimum air injection rates during drilling. Too much air caused higher pressures in the annulus and excessive erosion of surface equipment. Too little air caused inadequate hole cleaning and problems with low penetration rate, bit wear, and plugging.

FIELD TESTING

Three Pike County, Ky., wells were drilled with DFD techniques: COOP Well 1, COOP Well 2, and the experimental development (ED) well.

COOP Wells 1 and 2 were cooperative efforts with an Appalachian basin producer. In these wells, GRI drilling operations began after the operator's 8 5/8-in. intermediate casing string was set and ended after logging operations at total depth.

The ED well was permitted by RLG as operator for GRI and drilled and completed as a research well for GRI. Five test intervals were drilled with reverse circulation (Table 1). As each test interval was drilled, problems and successes were noted, and equipment and procedures were improved.

COOP WELL 1

The first reverse circulation test was started immediately below the 8 5/8-in. intermediate casing set in the top of the Mississippian Big Lime formation. Starting below this casing point eliminated the possibility of existing dust and cuttings from affecting the experiment. The interval reverse drilled was from 2,589 to 2,671 ft. A 7 7/8-in. hole was drilled with 4-in. drill pipe and 6 1/2-in. drill collars.

The only modifications made to normal drilling equipment were to install a kelly cock for flow control, install a minimal amount of fittings and valves on the existing standpipe and flow line to reverse the flow, and bore out the center of a tricone button bit to a 1 1/4 in. diameter (Fig. 2, Bit 1).

This first attempt was unsuccessful. The penetration rate was low, the bit life was short, and drill cuttings were not as large as desired, probably from regrinding of the cuttings because of poor hole cleaning. However, the absence of bit plugging problems and extreme surface equipment wear was encouraging. The bit needed further modifications to better clean the drill cuttings away from the bit.

The second test was in the Devonian shale from 4,405 to 4,657 ft. The only change was to the 7 7/8-in. bit. The jets were blanked off, a 1 1/2in. hole was bored in the center, and short, triangular skirts were installed on the blank jets between the cones (Fig. 2, Bit 2). Enlargement of the center bore was believed to result in better (and larger) cuttings returned because of less pressure, drop and, therefore, higher air velocity. Blanking the jets and installing the short skirts was thought to increase bit cleaning and cooling efficiency.

The results were encouraging. The sampling was improved with much larger cuttings and a shorter and more precise lag time for cuttings return. Although no significant shows were drilled, the BHTV revealed a very clean well bore, evidenced by the absence of dust and a clear view of well bore fractures.

However, the penetration rate was still substantially lower than that for regular drilling. The average penetration rate was 2.7 min/ft, including connection time, compared to a penetration rate of 0.7-1.0 min/ft with normal circulation in this same interval (reported by the same drilling contractor for several offset wells). The basic bit prior to modification and the operating parameters were the same as those in normal drilling. Thus, the target for further improvement was hole cleaning.

BETTER EQUIPMENT

The design and methodology for the next reverse circulation test were improved based on sequential equipment modifications and the experiences from the first two tests on COOP Well 1. The first concern was the safety of rig floor personnel during connections, the second was longer connection time, and the third was penetration rate during drilling.

The focus downhole was again on hole cleaning. As in the preceding test, the jets were blanked so air flow would be focused across the bit into the center bore, which was bored to 1 1/2 in. In addition, the skirts between each of the bit cones were extended downward and made an integral part of the bit (Fig. 2, Bit 3). The skirt extensions were designed to direct the air flow from the annulus across each cone of the bit. With the air channeling across the bit, bit cleaning and cooling should be enhanced, and the penetration rate should be increased.

During the first two tests, connection time was three to four times longer than that for normal circulation drilling. The annular air pressure had to be blown down before breaking the kelly to add a new joint. In addition, drill cuttings, dust, and any natural shows were exposed at the drill floor. This condition was both cumbersome and unsafe.

The solution was the use of a continuous circulating chamber (CCC) or air box (Fig. 3). The CCC is a split, cylindrical sleeve with hinges and latches for it to be placed across the kelly/drill pipe connection prior to separation. Seals on each end prevent escape of dust and cuttings, and a flow line connection to the blooie line provides a diverted path of discharge during the connection. A flapper-type check valve prevents discharge of dust and cuttings after the kelly has been removed.

The use of the CCC permits continuous circulation during the connection, minimizing the blowdown/pressure-up cycle which takes time and causes additional stresses on the well bore (Fig. 4).

Some other minor modifications, such as a separate standpipe and kelly hose dedicated to DFD operations, were also added to streamline the flow path.

COOP WELL 2

The third interval drilled with reverse circulation was the entire Devonian shale interval (3,585-4,487 ft) beginning just below the Berea sandstone and continuing to total depth. Because of the surface and downhole equipment changes, the penetration rate increased substantially. Table 2 compares similar shale intervals drilled with normal (COOP Well 1) and reverse (COOP Well 2) circulation:

The penetration rate with reverse circulation was as good as or better than that with normal drilling runs.
The bit life was as good as or better than that with normal drilling runs.
Less air rate was required for efficient hole cleaning and cuttings removal during reverse circulation.

In addition, the cuttings were drastically larger than that from normal drilling. Also, continuous circulation allowed better geologic depth control by eliminating cuttings fall back and contamination and reducing the lag time for cuttings return.

Some abrasive wear problems occurred in the surface equipment, but these problems were not considered major at this point. After completion of this work on COOP Well 2, the test had clearly demonstrated that reverse circulation could be used as a viable drilling technique.

Unfortunately, no gas shows were encountered; thus, there were no open hole permeability tests for skin measurements. The BHTV again revealed a very clean well bore.

ED WELL

The results of the previous tests indicated that reverse circulation drilling is applicable to drilling larger holes (that is, below the surface casing). The benefits include a decrease in the circulating air requirements, the introduction of annular back pressure to help reduce water influx, and help in minimizing pressure surges in the borehole.

On the experimental development well, the hole below the 11 3/4-in. surface casing was reverse-circulation drilled with an 11-in. bit (Fig. 2, Bit 4). This bit was virtually identical to the successful 7 7/8-in. bit from COOP Well 2, except for longer buttons and a 2-in. center jet. The drillstring had 4 1/2-in. drill pipe with 6 1/2-in. drill collars.

After the cement and float equipment were drilled out, the interval from 383 to 557 ft was reverse circulated, also using the air dryer. Although approximately 90 ft were drilled at better-than-normal penetration rates, plugging problems developed at the bit.

The hole also became damp, aggravating the problem. The exact cause of the plugging is uncertain, but suspicion was focused on the "dead space" between the cones and the center bore. During surface inspection of the plugged bit, a large accumulation of cuttings was discovered in the dead space, completely blocking the center bore. The large (1 in.) cuttings, a damp hole, and the dead space all probably contributed to the plugging problems.

After these conditions were fought for some time, the intermediate hole experiment was aborted. Normal circulation drilling with a soap mist continued to the intermediate casing point in the top of the Mississippian Big Lime formation at 2,533 ft. Once the 8 5/8-in. intermediate casing was cemented and drilled out, damage-free drilling (DFD) operations resumed with a 7 7/8-in. bit. Drilling continued from 2,564 to 3,582 ft, at which point DFD operations were aborted because of extreme surface equipment wear and the related expenses of replacement.

Several shows were encountered in this test interval, and numerous TWDs were conducted at various points to quantify the size and profile (sustained, declining, etc.) of the flows (Fig. 5). A staged drilling experiment helped determine the effects of drilling parameters on penetration rate (Table 3).

Other than the problems with abrasive surface wear, the DFD operations in this interval were very successful. Penetration rates were as good as or better than those with normal drilling, especially in the harder limestone in the Big Lime formation. Bit wear was negligible, lag time was short (40-45 sec with an air rate of 1,500-1,600 scfm), and cuttings were dramatically larger (Fig. 6). Also, connection time was reduced, and the blowdown/pressure-up cycle was minimized by using the CCC.

After a return to normal circulation, the well was drilled to 4,148 ft, where 245 ft of oriented core was taken. Normal circulation drilling then continued to total depth at 4,781 ft. At each stopping point (core point and total depth), the background gas from the Big Lime declined to negligible levels.

After flowing for several hours, the flow rates returned to previous levels, indicating drilling damage had indeed occurred.

RESULTS

Test-while-drilling results from the ED well strongly suggest drilling damage from normal techniques and an absence of damage from DFD operations (Fig. 5).
BHTV pictures show the superior condition of the well bore following damage free drilling (Fig. 7) compared to the same interval drilled with normal techniques (Fig. 8).
Open hole permeability tests prove the presence of damage to the well bore following normal air drilling, suggesting damage-free drilling may be useful.
Damage-free drilling provides superior samples with better depth control and less contamination. These samples can be used to obtain formation properties usually obtained from core analysis. The increase in cuttings size should permit testing for porosity, fluid saturations, and matrix permeability. The large cuttings may be suitable for preparation of thin sections.3
Damage-free drilling meets or exceeds penetration rates from normal drilling with equivalent drilling parameters.
Damage-free drilling can use lower air volumes than normal drilling uses, as demonstrated in the ED well. Reverse-circulation air drilling can be performed safely with the continuous circulating chamber and proper well control considerations.
Modifications to surface equipment are still required.

RECOMMENDATIONS

Excessive wear on surface equipment from drill cuttings moving at high velocity needs to be reduced for reverse circulation or DFD to become a commonly used technique. Different or modified equipment is needed to handle the air stream returns between the kelly and the blooie line. The existing equipment has too many turns and points of concentrated wear and is too small in diameter.

With current equipment, the cuttings travel vertically at 45 mph until they reach the gooseneck on top of the kelly. From there, the cuttings change direction 180, then travel through the kelly hose, make another 180 turn, continue through the standpipe connection making another 180 turn, and finally flow into the blooie line where an approximate 90-120 turn is made. All of these turns are points of concentrated wear.

RLG recommends general equipment changes to eliminate the inessential direction changes, reduce the velocity of the cuttings downstream of the top of the kelly, and provide a cushioning effect (possibly by the injection of water at or near the top of the kelly). A kelly spinner or top drive should be especially helpful. Additional research should focus on designing, building, and testing equipment to solve these problems and permit DFD to become a viable tool for air drilling low-permeability reservoirs.

REFERENCES

1. Reuben L. Graham Inc., "Exploration-Production Studies in Newly Drilled Devonian Shale Gas Wells," annual report to the Gas Research Institute, GRI-86/0119, February 1986.

2. Reuben L. Graham Inc., "Exploration-Production Studies in Newly Drilled Devonian Shale Gas Wells," quarterly report to the Gas Research Institute, November 1986.

3. Luffel, D.L., and Guidry, F.K., "New Core Analysis Methods for Measuring Reservoir Rock Properties of Devonian Shale," SPE paper 20571, presented at the 1990 SPE Annual Technical Conference and Exhibition, New Orleans, Sept. 23-26, 1990.