Method and system for determining joint quality of polycrystalline diamond (PCD) inserts to carbon steel bearing housings

Abstract

A method for evaluating the joint quality of poly crystalline diamond (PCD) inserts in a bearing assembly is disclosed. The method involves generating a set of reference images for a specific design of PCD bearing assembly based on ultrasonic testing results, where the reference images represent different percentage levels of low-quality bond. Statistical analysis is performed on the percentage levels of low-quality bond to establish pass / fail criteria. Ultrasonic testing is then performed on a PCD bearing assembly under test, generating a set of test images. The test images are compared to the reference images to determine a percentage level of low-quality bond for each test image. Based on the percentage level of low-quality bond for each test image, the PCD bearing assembly under test is determined to pass or fail according to the established pass / fail criteria. The method enables non-destructive evaluation of joint quality in PCD bearing assemblies.

Description

[0001] METHOD AND SYSTEM FOR DETERMINING JOINT QUALITY OF POLYCRYSTALLINE DIAMOND (PCD) INSERTS TO CARBON STEEL BEARING HOUSINGS

[0002] TECHNICAL FIELD

[0003] This disclosure is related to the testing of bearing assemblies, and in particular, to the non-destructive ultrasonic testing of bearing assemblies utilizing polycrystalline diamond (PCD) inserts.

[0004] BACKGROUND

[0005] Downhole drilling tools operate in harsh environments characterized by high shear forces, high pressure, rapid vibration, abrasive drilling muds, and high temperature. Components of these drilling tools, such as but not limited to steering units and power alternators, rely on bearings to provide control and stability. PCD bearing assemblies have become increasingly utilized in these applications due to their significantly longer lifespan compared to conventional roller or plain bearings. PCD bearing assemblies can be configured in various geometries, including thrust and radial (male and female). For example, a male radial bearing can be positioned within a female radial bearing, with a thrust bearing positioned between the two to accommodate axial loads.

[0006] PCD is a synthetic diamond material produced by sintering small diamond crystals under high pressure and temperature in the presence of a metal catalyst (onto a substrate). The resulting material exhibits a combination of extreme hardness, high thermal conductivity, high strength, high toughness, and low friction, making it particularly well-suited for use in harsh downhole drilling environments.

[0007] In typical PCD bearing assemblies 10, as illustrated in FIGS. 1A, IB, and 1C, small PCD inserts 12 are affixed within cavities in body 11 (often steel rings or housings or bodies). These inserts are secured using a brazing process, creating a brazed joint (shown and described in FIG. 2 as 14a) that is critical to the bearing's performance. In particular, shown in FIG. 1A is a male bearing assembly 10, shown in FIG. IB is a female bearing assembly 10a, and shown in FIG. 1C is a first half of an axial bearing assembly 10’. Although, the second half of the axial bearing assembly 10’ is not shown, it looks similar, with the PCD inserts of the second half facing the PCD inserts 12 of the first half.

[0008] FIG. 2 provides a diagrammatic cross-sectional view of this arrangement, in which a PCD insert 12 is shown disposed within a cavity or pocket 11b formed in the body 11 of a bearing assembly. The PCD insert 12 itself comprises a tungsten carbide substrate 12a with a polycrystalline diamond (PCD) table 12b formed on the top surface thereof. Brazing material 14 bonds the PCD insert 12 to the bottom surface and sidewall of the cavity 1 lb, enabling load transfer between the insert and the body 11 forming a brazing joint 14a.

[0009] Despite the inherent wear resistance of PCD, premature failures of PCD bearings have been observed during downhole drilling operations. These failures occur earlier and more frequently than expected and are characterized by the detachment and liberation of the PCD inserts 13 from the steel housings 11 and 1 la as depicted in FIGS. 3A and 3B. Examination of failed bearings has revealed that the brazing joints 14a represent the weakest point, with the PCD inserts detaching intact.

[0010] In particular, detailed inspections have shown that brazing defects leading to cracking, voids, delamination, and / or detachment frequently initiate at the interface between the brazing material 14 and the metal substrate (e.g., the body 11 or the tungsten carbide substrate 12a of the PCD insert). Thermal stresses are believed to be a primary cause of these cracks. Indeed, joint quality (e.g., strength) of the brazed joint 14a between the PCD insert 12 and the walls / floor of its cavity 1 lb, which is clearly reduced by such cracks, has been identified as a critical factor of the performance and longevity of PCD bearing assemblies.

[0011] FIG. 4 illustrates the crucial relationship between joint strength and the load energy (both mechanical and thermal) encountered during drilling operations. Shown in which would result in detachment of the PCD inserts are graphs of a probability distribution over load energy encountered during drilling and graphs of a probability distribution over acceptance energy for the strength of the PCD braze joints of PCD bearing assemblies. The x-axis of the graphs represent an energy value corresponding the load acting on a joint or bond of a PCD bearing assembly. The x-axis also represents the energy corresponding to the strength of a joint or bond of a PCD bearing assembly. The PCD bearing assembly including multiple PCDs with multiple joints or bonds. The y-axis of the graphs represent a probability that a joint or bond is encountering a respective load energy. The y-axis also represents the probability that a joint or bond has a respective strength (acceptance energy).

[0012] First consider the illustrated "high energy condition", representing a situation where the provided mechanical or thermal energy partially exceeds the strength of the joint or bond. The overlap between the energy and strength curves in this condition signifies an elevated risk of failure. When the energy surpasses the joint's strength, even briefly, it can cause the detachment of the PCD inserts, compromising the bearing's integrity and functionality.

[0013] Next, a "low strength condition" is depicted, which represents a situation where the initial strength (to) of the joint or bond is lower than expected, resulting in an inadequate safety margin between the strength and provided energy. Despite the mechanical and thermal energy levels remaining within the designed range, the diminished strength renders the joint or bond vulnerable to failure when subjected to the anticipated energy levels.

[0014] A "decayed strength condition" is also illustrated, demonstrating the excessively fast deterioration of the joint strength over time (to to ti) due to wear and tear. Initially, the joint or bond strength aligns with the desired level, providing an adequate safety margin. However, as the component undergoes stress and exposure to the demanding downhole environment, the strength progressively declines at a high rate. Eventually, the strength dips below the provided load energy curve, resulting in an increased likelihood of failure.

[0015] Given this, it is clear that an adequate safety margin where the strength of the PCD braze joint consistently surpasses the provided energy throughout the component's life is desirable - this condition is shown as the "desired design condition", as it helps ensure that the joints can withstand the mechanical and thermal energy encountered during drilling operations without failing prematurely, which would result in detachment of the PCD inserts.

[0016] Ensuring that this adequate safety margin is met requires addressing two primary challenges. First, the production of PCD braze joints must consistently achieve the required strength. Second, inspection methods are needed to verify joint quality and ensure that the necessary safety margin for PCD braze joints is met in PCD bearing assemblies before they are deployed in a drill string. Regarding the production of high-quality brazed joints, the brazing process itself can be particularly problematic. This is especially true when PCD elements (particularly the tungsten carbide substrate 12a supporting the PCD table 12b) are brazed into pockets 1 lb in the body 11 when the cooling rate after brazing is uncontrolled, which can lead to cracks, delamination, and lateral voids (collectively referred to here as “joint defects” or “brazing defects”).

[0017] Therefore, thorough inspection of joint quality is essential to identify cracked or otherwise deficient brazed joints, ensuring that the required safety margin is met by a given PCD bearing assembly before it is used in a drill string. However, non-destructive inspection of this joint quality presents challenges due to the interface between the PCD insert 12 and the bearing housing 11. While destructive methods like sectioning and metallography can reveal joint defects, these techniques are costly, time-consuming, and limited to a small number of samples.

[0018] Traditional radiographic testing techniques, commonly employed for non-destructive evaluation, encounter significant limitations when applied to PCD bearing assemblies. These limitations arise due to the presence of tungsten carbide (12a) in the PCD inserts. Tungsten is a dense element with a high atomic number, making it an effective absorber of X-rays and gamma rays. This high absorption results in reduced penetration of the radiation. When X-rays or gamma rays encounter the tungsten carbide substrate, a significant portion of the radiation is absorbed, reducing the amount that penetrates through the braze joints to be inspected and reaches the detector. The resulting image has poor contrast and low sensitivity in the region of the braze joint, making it difficult to observe small defects or variations in braze material density that are indicative of a poor braze joint. Thus, in essence, the tungsten carbide acts as a shield, preventing proper inspection of the braze joint.

[0019] While ultrasonic testing presents a viable alternative to radiographic methods, its application to PCD bearing assemblies has been limited by several significant challenges. Prior attempts at ultrasonic inspection of these components have suffered from the absence of standardized reference images or data thresholds that correlate ultrasonic results with actual joint quality. Without established calibration protocols specific to PCD insert joints, test-to-test variance has been considerable, resulting in inconsistent evaluations of the same or similar joints. Furthermore, the industry has lacked a systematic methodology for analyzing ultrasonic data and establishing statistically sound pass or fail criteria based on known failure patterns. Traditional ultrasonic approaches have also failed to adequately address the relationship between joint strength and the mechanical / thermal energy encountered during drilling operations. These limitations have hindered the widespread adoption of ultrasonic testing for PCD bearing assemblies despite its potential advantages, leaving the industry without reliable non-destructive means to assess joint quality prior to field deployment and prevent premature failures.

[0020] It follows then that given the current insufficient testing techniques, the oilfield industry has experienced premature bearing failures in the field. This is rather undesirable. Therefore, to ensure the reliable performance of PCD bearings and avoid costly downhole failures that result in tool repairs and lost drilling time, there is a clear need for improved non-destructive methods for assessing the quality of these joints.

[0021] SUMMARY

[0022] A method of evaluating a joint of a PCD insert in a PCD product includes obtaining, using an ultrasonic testing apparatus, a set of reference data of a reference PCD product. The reference PCD product includes a plurality of reference PCD inserts in a plurality of reference pockets positioned in a reference support structure, with the plurality of reference PCD inserts joined to the reference support structure by a plurality of reference bonds. The method further includes calibrating the ultrasonic testing apparatus using the set of obtained reference data and obtaining, using the calibrated ultrasonic testing apparatus, a set of measurement data of a sample PCD product. The sample PCD product includes a plurality of sample PCD inserts positioned in a plurality of sample pockets in a sample support structure, with the plurality of sample PCD inserts joined to the sample support structure by a plurality of sample bonds. The method also includes defining a pass or fail criterion for the sample PCD product, analyzing, using a processor, the set of measurement data and applying the pass or fail criterion, and operating the sample PCD product in a wellbore when the analyzed set of measurement data meets the pass criterion. The set of measurement data may be based on reflections of an ultrasound wave emitted by the calibrated ultrasonic testing apparatus at an interface between the sample bond and the sample support structure.

[0023] Defining a pass or fail criterion may include defining multiple pass or fail criteria.

[0024] Obtaining a set of reference data may include using the ultrasonic testing apparatus on a reference PCD product with reference bonds of known bond quality.

[0025] Analyzing the set of measurement data may include defining a threshold, and then determining whether each item of measurement data in the set of measurement data is above or below the threshold.

[0026] Each sample bond in the plurality of sample bonds may be represented by a subset of measurement data of the set of measurement data, and the analyzing may include determining how many measurement data in each subset of measurement data are above the threshold.

[0027] The pass or fail criterion may define how many of the measurement data of each subset of measurement data are allowed to be above the threshold, and the analyzing may include rating each of the sample bonds of the plurality of sample bonds as passed or failed based on the pass or fail criterion.

[0028] Analyzing the set of measurement data may include defining a marginal band, and then determining whether each item of the measurement data of the set of measurement data is within the marginal band.

[0029] The pass or fail criterion may define how many sample bonds of the plurality of sample bonds are allowed to be within the marginal band while still maintaining operability of the sample PCD product.

[0030] Obtaining the set of measurement data may include transmitting an ultrasound wave having a first amplitude and receiving a reflected ultrasound wave having a second amplitude, determining a ratio of the first amplitude and the second amplitude, and using the ratio to analyze the set of measurement data.

[0031] The PCD product may be a bearing assembly.

[0032] Obtaining, using the calibrated ultrasonic testing apparatus, a set of measurement data of a sample PCD product may include transmitting an ultrasound wave from a transducer, where the ultrasound wave first passes a portion of the sample support structure and then enters a sample bond of the plurality of sample bonds.

[0033] The method may further include obtaining multiple sets of measurement data from multiple sample PCD products, with each sample PCD product being evaluated according to the method described above. The method may also include storing the multiple sets of measurement data in a database, obtaining an additional set of measurement data from an additional sample PCD product using the calibrated ultrasonic testing apparatus, updating the pass or fail criterion based on the additional set of measurement data, and re-analyzing the multiple sets of measurement data stored in the database based on the updated pass or fail criterion.

[0034] Updating the pass or fail criterion may include updating a threshold.

[0035] Analyzing the set of measurement data may include using an image analysis software.

[0036] Calibrating the ultrasonic testing apparatus may include obtaining a set of reference data that includes more than 50 reference data points.

[0037] Calibrating the ultrasonic testing apparatus may include setting one of a sensitivity or a gain of the ultrasonic testing apparatus.

[0038] A system for evaluating a joint of a PCD insert in a PCD product includes an ultrasonic testing apparatus configured to obtain a set of reference data of a reference PCD product. The reference PCD product includes a plurality of reference PCD inserts positioned in a plurality of reference pockets in a reference support structure, with the plurality of reference PCD inserts joined to the reference support structure by a plurality of reference bonds. The ultrasonic testing apparatus is further configured to be calibratable based upon the set of obtained reference data and to obtain a set of measurement data of a sample PCD product. The sample PCD product includes a plurality of sample PCD inserts positioned in a plurality of sample pockets in a sample support structure, with the plurality of sample PCD inserts joined to the sample support structure by a plurality of sample bonds. The system also includes a processor configured to define a pass or fail criterion for the sample PCD product, analyze the set of measurement data and apply the pass or fail criterion, and operate the sample PCD product in a wellbore when the analyzed set of measurement data meets the pass criterion. The set of measurement data may be based on reflections of an ultrasound wave emitted by the calibrated ultrasonic testing apparatus at an interface between a given sample bond of the plurality of sample bonds and the sample support structure.

[0039] The processor may be configured to define multiple pass or fail criteria.

[0040] The ultrasonic testing apparatus may be configured to obtain the set of reference data using a reference PCD product with reference bonds of known bond quality.

[0041] The processor may be configured to analyze the set of measurement data by defining a threshold, and then determining whether each item of measurement data in the set of measurement data is above or below the threshold.

[0042] Each sample bond in the plurality of sample bonds may be represented by a subset of measurement data of the set of measurement data, and the processor may be configured to determine how many measurement data in each subset of measurement data are above the threshold.

[0043] Analyzing the set of measurement data may include defining a marginal band, and then determining whether each item of the measurement data of the set of measurement data is within the marginal band.

[0044] The pass or fail criterion may further define how many of the sample bonds of the plurality of the sample bonds in the PCD product are allowed to be rated as failed while still indicating the sample PCD product is ready for use.

[0045] The pass or fail criterion may define how many sample bonds of the plurality of sample bonds are allowed to be within the marginal band while still indicating the sample PCD product is ready for use.

[0046] The ultrasonic testing apparatus may be configured to obtain the set of measurement data by transmitting an ultrasound wave having a first amplitude and receiving a reflected ultrasound wave having a second amplitude, determining a ratio of the first amplitude and the second amplitude, and using the ratio to analyze the set of measurement data.

[0047] The PCD product may be a bearing assembly.

[0048] The ultrasonic testing apparatus may include a transducer configured to transmit an ultrasound wave, where the ultrasound wave first passes a portion of the sample support structure and then enters a sample bond of the plurality of sample bonds. The ultrasonic testing apparatus may be configured to obtain multiple sets of measurement data from multiple sample PCD products. The system may further include a database configured to store the multiple sets of measurement data. The ultrasonic testing apparatus may be further configured to obtain an additional set of measurement data from an additional sample PCD product, and the processor may be configured to update the pass or fail criterion based on the additional set of measurement data and re-analyze the multiple sets of measurement data in the database based on the updated pass or fail criterion.

[0049] The processor may be configured to update the pass or fail criterion by updating a threshold.

[0050] The processor may include an image analysis software configured to analyze the set of measurement data.

[0051] The ultrasonic testing apparatus may be configured to be calibrated using a set of reference data that includes more than 50 reference data points.

[0052] The ultrasonic testing apparatus may be configured to be calibrated by setting one of a sensitivity or a gain of the ultrasonic testing apparatus.

[0053] BRIEF DESCRIPTION OF THE DRAWINGS

[0054] FIGS. 1A-1C are diagrammatical representations of bearing assemblies including polycrystalline diamond (PCD) inserts.

[0055] FIG. 2 is a cross-sectional view of a PCD insert within a cavity in a body of a bearing assembly.

[0056] FIGS. 3A and 3B are diagrammatical representations of bearings including PCD inserts that have failed (detached) during operation.

[0057] FIG. 4 is a graph of probability occurence for mechanical / thermal energy encountered during drilling and proability occurrence for the strength of the PCD joints.

[0058] FIG. 5 is a diagrammatical representation of a bearing assembly including PCD inserts that have failed during operation, undergoing ultrasonic testing by an ultrasonic testing apparatus shown as a block diagram. FIG. 5A is a cross-sectional view of a PCD insert within a cavity in a housing of a bearing assembly undergoing ultrasonic testing.

[0059] FIG. 5B is a cross-sectional view of a portion of a first bearing assembly with multiple PCD inserts therein, undergoing ultrasonic testing.

[0060] FIG. 5C is a cross-sectional view of a portion of a second bearing assembly with multiple PCD inserts therein, undergoing ultrasonic testing.

[0061] FIG. 5D is a cross-sectional view of a portion of a third bearing assembly with multiple PCD inserts therein, undergoing ultrasonic testing.

[0062] FIG. 6 shows ultrasound data of the cavity / PCD insert combination of the bearing assembly of FIG. 5, taken prior to the placement of the bearing into operation.

[0063] FIG. 7 schematically illustrated the results of destructive testing of a bearing assembly including a PCD insert that has failed during operation, as well as the ultrasound data taken of that cavity / PCD insert combination prior to placement of the bearing into operation, to as to show verification of the correlation of the ultrasound data to bond quality and bearing failure.

[0064] FIGS. 8-9 show the correlation of images of a set of reference ultrasound images of cavity / PCD insert combinations to the percentage of the joints between the sidewall and floor of the cavity and the PCD insert that are of poor quality.

[0065] FIGS. 10-11 are diagrammatical representations of a set of reference images as correlated to joint quality.

[0066] FIG. 12 shows ultrasound data of the cavity / PCD insert combinations of a bearing assembly including PCD inserts that have failed during operation, as well as the statistical analysis thereof performed in order to determine pass / fail conditions.

[0067] FIG. 13 shows the raw ultrasound data of a cavity / PCD insert combination prior to placement of the bearing into operation.

[0068] FIG. 14 is a block diagram of an apparatus for performing the ultrasonic testing techniques of this disclosure.

[0069] DETAILED DESCRIPTION

[0070] The following disclosure enables a person skilled in the art to make and use the subject matter described herein. The general principles outlined in this disclosure can be applied to embodiments and applications other than those detailed above without departing from the spirit and scope of this disclosure. It is not intended to limit this disclosure to the embodiments shown, but to accord it the widest scope consistent with the principles and features disclosed or suggested herein.

[0071] As previously explained, a non-destructive testing technique is desired to inspect the joints (braze joints, as a non-limiting example) of PCD inserts in bearing assemblies and ensure that the joints are in the desired condition with a safety margin and will not be expected to fail within the other described conditions during the desired lifetime. To that end, ultrasonic testing is a non-destructive testing method that may be used to detect internal flaws, measure thickness, and characterize materials. This disclosure presents techniques for performing ultrasonic testing on the joint between PCD inserts and the bearing housing, as well as for the calibration and interpretation of the test results, in order to provide a non-destructive test of the joint quality. The bearing housing is also referred to herein as support structure or body.

[0072] Refer additionally to FIG. 5, showing a diagrammatical depiction of a bearing assembly 10 having two rows Rowl, Row2 of cavities formed about the outer diameter of the body 11 of the bearing assembly, with each cavity intended to have a PCD insert 12 bonded thereinto. The cavities in FIG. 5 are all shown as occupied by a PCD insert. Rowl shows cavities (each with a respective PCD insert therein), with labels starting with 1-1, and Row2 shows cavities (each with a respective PCD inserts therein), labeled with 2-1. The amount of cavities or PCD inserts is set per product size, function and application. Rowl and Row 2 are therefore intended to have the exact amount of PCD inserts to fill the cavities. In FIG. 5, however, the bearing assembly is a unit that has failed prematurely, with all inserts in the first row Rowl having fallen out, here labeled as detached PCD inserts 13.

[0073] Ultrasonic testing may be performed on this bearing assembly by transducing from the inner diameter of the steel body of the bearing assembly, using an ultrasonic testing apparatus 100. To that end, the ultrasonic testing apparatus 100 utilizes an ultrasonic transducer 15, which generates high-frequency sound waves 16 (e.g. 10 MHz) and receives the reflected sound waves. The sound waves are also referred to herein as ultrasonic waves or ultrasound waves. In the case of a bearing assembly for a drill string, the transducer 15 of the ultrasonic testing apparatus 100 is typically mounted on ahead 17 fixed to a rotating carrier 18 that can be inserted into the bore 19 of the bearing assembly 10, although a head 17 rotatingly mounted on a fixed carrier 18 may also be utilized. This allows the transducer 15 to scan the entire circumference of the bearing assembly as it moves along the axial direction along axis Al, ensuring full coverage of all the PCD inserts 12 on the bearing assembly. The transducer 15, supporting head 17, and carrier 18 are configured to scan 360 degree of the circumference of the bearing assembly 10 and at least an axial extension that includes all PCD inserts 12 (and detached PCD inserts 13) in the two rows, Rowl and Row2.

[0074] The ultrasonic testing apparatus 100 includes a power supply 22, a measurement unit 23, a signal generator 24, a processing unit 25, a memory and / or a database 28, cabling 26, and a display 27. The power supply 22 provides power to the transducer 15 and the electronic components (22, 23, 24, 25, 27, 28) of the apparatus 100. The measurement unit 23 controls the ultrasonic measurement by defining parameters such as signal frequency, signal intensity, measurement sensitivity, measurement timing, and the application of calibration data (e.g., sensitivity, gain). The signal generator 24 produces an ultrasound signal for the transducer 15 to transmit ultrasonic waves. The processing unit 25 processes and / or analyzes the measured data, while the memory and / or database 28 stores the measured and / or analyzed data. The display 27 presents the measured and / or analyzed data for review.

[0075] The transducer 15 transmits sound waves that are coupled to the inner diameter surface of the bearing assembly 10 using a couplant 18, such as a specialized fluid (e.g., water), to facilitate the transmission of ultrasonic waves into the material. The wavelength of the ultrasonic waves is selected to penetrate the body 11 of the bearing assembly 10 and interact with the joint 14a (FIG. 5 A). The beam size is configured to provide adequate resolution for inspecting the joint area and determining the size and perimeter of the joint area to evaluate the bond between the PCD inserts 12 or 13 and the body 11 or support structure of the bearing assembly 10.

[0076] FIG. 5 A provides a zoomed-in view of the pathway of the sound wave 16. The sound waves transmitted by the transducer propagate through the body 11 and enter the joint 14a. A dashed line indicates potential reflection planes, which may be defined by material changes or defects in the joint 14a. After propagating through the body 11 and the joint 14a, the ultrasonic waves 16 are reflected back to the transducer 15 from interfaces between different materials (e.g., steel / filler material, filler material / PCD) and from any defects or discontinuities in the joint 14a. The filler material may be brazing material, welding material, vapor deposition material, or adhesive material. The reflected waves are converted into electrical signals by the transducer 15 and analyzed to characterize the quality of the joint 14a. The ultrasonic waves may propagate perpendicular to the joint 14a orientation or perpendicular to the floor plane of the cavity 1 lb. In an alternative embodiment the ultrasonic wave may propagate at an angle different to 90 degrees (not shown) to the floor plane of the cavity 1 lb.

[0077] As an alternative, ultrasonic testing may be performed by coupling the transducer to the outer diameter surface of the bearing assembly or by using a transmission setup where a transducer is positioned on one side of the outer diameter and a receiver on the opposite side. This setup is particularly useful when the PCD inserts are joined to the inner diameter of the bearing assembly 10, such as in a female bearing assembly, as opposed to a male bearing assembly shown in FIG. 5.

[0078] FIGS. 5B-5D illustrate different setups of the testing apparatus 100 based on the type of bearing assembly 10 being evaluated. Specifically, FIG. 5B depicts the setup for a male radial bearing assembly 10, FIG. 5C depicts the setup for a female radial bearing assembly 10a, and FIG. 5D depicts the setup for a first half of an axial bearing assembly 10’, also referred to as a thrust bearing. FIGS. 5B-5D indicate the path of the sound wave from the transducer 15 to the joint 14a and the path of the reflected soundwave from the joint to the transducer. The soundwave passes through the couplant 18 (FIG. 5). In FIGS. 5B-5D the couplant is assumed to be water. The distance between the transducer 15 and the joint 14a in FIGS. 5B and 5C is around 5 mm (water path (WP) ~ 5mm). In FIG. 5D the distance is 15 mm (water path (WP) ~15 mm).

[0079] In greater detail, FIG. 5B shows a bearing surface 30 on the male radial bearing assembly 10 that corresponds to a bearing surface30a on the female bearing assembly 10a as displayed in FIG. 5C. When the male bearing assembly 10 and the female bearing assembly 10a are combined, they form a full radial bearing assembly, where one part (male or female) serves as the stator and the other as the rotor. During operation, the bearing surfaces 30 and 30a slide against each other. Although the second half of an axial bearing assembly is not depicted, it is similar to the described configuration.

[0080] The results of the ultrasonic testing for each joint or each cavity / PCD insert combination of the bearing assembly shown in FIG. 5 — prior to being placed into service, and therefore prior to failure — are depicted in FIG. 6. The image for each PCD insert ranges in shade from light (green, in a colorized image) to dark (red, in a colorized image). Once placed into service, the bearing assembly failed, with all of the inserts in the first row, Rowl, having fallen out.

[0081] In FIG. 6, the average shade of each cavity in Rowl is substantially darker (containing substantially more red in a colorized version) than the average shade of each cavity in Row2 (which contains substantially more yellow in a colorized version). The darker color indicates a greater amplitude of the reflected sound wave, as joint defects lead to stronger reflections. This demonstrates a correlation between the ultrasonic test results performed on the bearing assembly after manufacture but prior to being placed into service and the occurrence (and location) of early PCD insert failures.

[0082] Each pixel in FIG. 6 represents one measurement value corresponding to one location of the joint (e.g., brazing bond), specifically a location within the floor of the cavity where the PCD insert is situated. The resolution (i.e., the size of a pixel in the image) of a typical ultrasonic image may be smaller than 0.1 mm, smaller than 0.5 mm, or smaller than 1 mm. A typical diameter of a PCD insert ranges between 5 mm and 20 mm.

[0083] Confirmation of this correlation will be discussed hereinbelow, but first, the failure mode for the PCD inserts in Rowl will be discussed in more detail. When the joint between the sidewalls / floor of a cavity and a PCD insert fails, the PCD insert loses its secure attachment to the bearing assembly. This allows the PCD insert to begin to rotate or spin within the cavity during operation, as it is no longer held in place by the joint (weakend joint). The spinning motion of the PCD insert causes it to wear against the sidewalls and the floor of the cavity, enlarging the cavity and further compromising the fit between the insert and the cavity. As the cavity becomes larger and the PCD insert becomes looser, the rate of rotation increases, accelerating the wear process. Eventually, the PCD insert may become completely detached from the cavity and fall out of the bearing assembly during operation. The loss of one or more PCD inserts from the bearing assembly leads to a significant imbalance in the bearing assembly, as the remaining inserts are subjected to increased loads and vibrations. This imbalance can cause accelerated wear and damage to the adjacent PCD inserts, leading to a cascading failure mode where the loss of one PCD insert quickly leads to the failure of others. This domino effect is likely responsible for the complete loss of all PCD inserts in Rowl of the bearing assembly shown in FIG. 5.

[0084] The described failure mode highlights the critical importance of ensuring adequate joint quality in PCD bearing assemblies. A strong, defect-free joint is necessary for retaining the PCD inserts in their respective cavities and preventing the onset of the spinning, wearing, and cascading failure process. By using ultrasonic testing to non-destructively evaluate the joint quality prior to placing the bearing assembly 10 into service, potential issues with the bearing assembly can be identified and the bearing assembly either repaired or discarded. In case the joint quality of the PCD bearing assembly is evaluated to be of good quality, the PCD bearing assembly can be operated in a wellbore operation, wherein operation includes the provision of the PCD bearing assembly for a wellbore operation, such as by delivering or selling the PCD bearing assembly.

[0085] Returning to the correlation between the ultrasonic test results performed on the bearing assembly after manufacture but prior to being placed into service and whether (and where) an early PCD insert failure occurs, confirmation has been obtained through destructive testing. FIG. 7 illustrates the results of destructive testing of a bearing assembly, including a PCD insert that has failed during operation, as well as a PCD insert that remains within its cavity in the bearing assembly. The figure also shows the ultrasonic testing results for the cavity / PCD insert combination (joint) that failed and a schematic depiction of a cross-section through the cavity / PCD insert combination.

[0086] In the darkest (reddest in a colorized version) shaded areas, insufficient adhesion of the filler material (e.g., brazing material) to either the tungsten carbide substrate of the PCD insert or the steel body of the bearing assembly can be observed. This insufficient adhesion is primarily caused by cracks and voids. These observations confirm the correlation between ultrasonic test results and joint quality. Note that the filler material is also referred to herein as joint material. A workflow used to determine testing criteria and to evaluate the joint quality of PCD inserts in bearing assemblies is now described.

[0087] First, a set of reference images is generated for each specific design of PCD bearing assembly (such as the male bearing assembly of FIG. 5) that will undergo ultrasonic testing to assess the joint quality. For example, if three different PCD bearing assembly types (e.g., three different part numbers, or more generally, three different types of PCD bearings, such as radial male, radial female, and thrust) are to be evaluated, then three sets of reference images are created. In the set of a reference image are images for each joint, as shown in FIG. 6.

[0088] These reference image sets are based on ultrasonic testing results obtained from PCD bearing assemblies. Refer to the examples shown in FIGS. 8-9, which present enlarged images representing the ultrasonic testing results for two cavity / PCD insert combinations (joints) in a PCD bearing assembly known to have experienced failure (also referred to herein as a reference PCD product). Destructive testing was performed to determine the actual quality of the joints between the PCD inserts and the sidewalls / floors of the cavities. In FIG. 8, it was found that 48% of the pixels of the image of the respective joint indiacte low joint quality, meaning that 48% of the joint was not effectively (or not at all) contributing to a sturdy bond. Similarly, in FIG. 9, 13% of the pixels of the image of the resecptive joint was found to be of low joint quality.

[0089] From a sufficient number of such examples for each PCD bearing design, a set of reference images is created. FIGS. 10-11 show a sample set of schematic reference images for a given reference PCD bearing assembly, representing the ultrasonic testing results for joints between the PCD insert and the support structure with different levels of quality content (e.g., 5% of the joint are of low quality, 10% of the joint are of low quality, up to 80% of the joint are of low quality, etc.). This is, FIG. 10-11 shows examples for a specific percentage of all pixels in the ultrasonic image of a joint (e.g. brazing bond) of one PCD insert that are identified to showing a bad bond between the PCD insert and the support structure of the floor or sidewall of the support structure.

[0090] Using these reference image sets along with the knowledge of which images correspond to failed joints (bad bonds), statistical analysis can be performed. FIG. 12 illustrates ultrasonic test results for a reference PCD bearing assembly (also referred to herein as a reference PCD product) known to have failed, with each PCD insert in Rowl having fallen out. The figure also shows the percentage of low-quality bond (low-quality joint) pixels (also referred to hering as area fraction) associated with each PCD insert based on a comparison to schematic reference images. From the ultrasonic test results, it is visually apparent that the average shade of each image in Rowl is substantially darker (containing more red in a colorized version) than the average shade of each image in Row2 (containing more yellow in a colorized version). The percentage data in the table indicates how many of the measurement data (image pixels), belonging to one bond (joint), have values above a defined threshold. A higher percentage value corresponds to worse bond quality for a specific joint. For example, the bond of cavity 1-4 (Rowl, cavity 4) has the worst quality among all the bonds in the bearing assembly 10, with 67% of the measurement data (image pixels) for cavity 4 in Rowl exceeding the threshold value. The table also shows the average percentage of all bonds of all cavities in a row, as well as the minimum and maximum percentage values within a row.

[0091] This analysis not only reveals the percentage of image pixels in a joint that have low- quality bond at which a specific joint can be expected to fail but also provides statistics indicating the likelihood of failure in a specific row. These statistics are particularly useful when considering the cascading failure mode described earlier, where the loss of a single PCD insert leads to the failure of adjacent inserts in the same row. For example, a row with a higher average percentage (low-quality bond) but a lower maximum percentage of the image pixels may be more likely to fail than a row with a lower average but a higher maximum percentage, or vice versa. Based on these findings, pass or fail criteria can be established. For instance, an inspected PCD bearing assembly (also referred to herein as sample PCD product or PCD product) may be considered unsuitable for use if it has one single percentage value greater than or equal to 25%, three or more conti guous / adjacent inserts with percentage values greater than 20%, or a row with an average percentage greater than 19%. In addition, the pass or fail criterion may comprise a single criterion or multiple pass or fail criteria that are evaluated simultaneously or sequentially. For example, a first pass or fail criterion may define a threshold percentage of low-quality bond pixels allowable for individual joints (individual PCDs), while a second pass or fail criterion may define the maximum number of adjacent failed joints permitted in a row, and a third pass or fail criterion may specify the overall percentage of failed joints allowable across the entire bearing assembly. The multiple criteria work together to provide a comprehensive evaluation framework that accounts for different failure modes and risk factors. Gathering more and more measurement data and storing them in a database provides the opportunity to use machine learning algorithms to look for pattern in the measurement data that can be associated with specific failure modes and that allow to refine pass or fail criteria such as threshold(s).

[0092] With the set of reference images corresponding to the PCD bearing assembly type of interest (same PCD product), ultrasonic testing (the calibration of which is discussed further below) is performed on a PCD bearing assembly ofthat type prior to it being placed into service. For each cavity / PCD insert combination (joint) in the resulting ultrasonic test results, the reference image set is used to determine the percentage of low-quality bond in the joint. The previously described statistical calculations are then performed, and the presence (or absence) of any failure conditions is determined. If a failure condition is identified, the PCD bearing may be discarded or repaired. Identifying a failure condition includes applying a pass or fail criterion. A fail criterion may specify that a certain percentage of the image pixels for one bond exceeds a predefined value, such as 30% of all image pixels for one bond being above the threshold. Another fail criterion may involve a specific number of conti guous / directly adjacent PCD inserts in the PCD bearing assembly with bonds rated as failed because the percentage of image pixels exceeds the threshold. The specific number may be 2, 3, or 4. The image pixels or measurement data belonging to a specific bond are referred to as a subset of measurement data, while all measurement data for one bearing assembly, such as a male PCD bearing assembly

[0093] 10, are referred to as a set of measurement data. In an embodiment the set of measurement data may not only include measurement data from the joints but as well from the portion of the body

[0094] 11, not including PCDs and that are without j oints, such as the portions of the body 11 that are between the PCD inserts or between the joints. These measurement data can be identified and be filtered out, e.g., by an amplitude gate or by a suitable time window in which the measurement unit will look for reflected sound waves.

[0095] In an alternative embodiment, instead of using reference images and instead of analyzing image pixel data, the quality of the joints in PCD bearing assemblies can be evaluated using raw ultrasound data detected by the transducer, such as a percentage of the full screen height of the apparatus - relative to a known (reflected) amplitude (full screen height may be a voltage value). In an alternative embodiment the raw ultrasound data is transfered into decibels (dB) values (such as relative to the emitted ultrasound signal strength). In one more embodiment the raw data is the measured valotage value of the transducer including calibration and / or data conditioning. This approach eliminates the need for generating and comparing reference images, as the raw ultrasound data is used directly for analysis.

[0096] In this method, a sample set of raw ultrasound data from different PCD insert joints of PCD bearings known to have failed is assembled. The raw ultrasound data represents the strength of the reflected ultrasound signal at different points along the joint interface. Sample raw ultrasound data for a cavity / PCD insert combination is shown in FIG. 13 (percentage of the full screen height). Here, the raw ultrasound data is superimposed over the colorized image formed therefrom to illustrate that higher values in dB indicate areas of poor bonding while lower values in dB indicate areas of good bonding.

[0097] To establish the relationship between the raw ultrasound data and the joint quality, destructive testing is performed on the PCD bearing assemblies from which the raw ultrasound data was collected. This allows for a direct comparison between the ultrasound data and the actual quality of the joints, as determined through physical examination. From this, a set of threshold values (in dB) can be established to categorize the joint quality. For example, ultrasound data below 25 dB within a given area may indicate a high-quality joint, while data above 25 dB within the given area may suggest a low-quality or compromised joint.

[0098] Using these threshold values, statistical analysis can be performed. This analysis can include calculating the average, minimum, and maximum dB values for each row of PCD inserts. Similar to the reference image approach, a pass or fail criterion or pass or fail criteria can be determined based on the statistical analysis of the raw ultrasound data. For example, a PCD bearing assembly may be considered unsuitable for use if it has one PCD insert with dB values in a given area above the threshold value, three or more conti guous / adjacent PCD inserts with dB values in a given area greater than the threshold value (or a different threshold value), or a row of PCD inserts with average dB values greater than the threshold value (or a different threshold value). The pass or fail criterion or criteria may consider beside the number of measurement data above the threshold (fail band) and below the threshold (pass band), measurement data in a marginal band which are contributing in the pass or fail decision based on other conditions. For example, a PCD bearing assembly may not show fail band measurement data or fail band image pixels, but may show three neighboring joints that show marginal band measurement data or marginal band imgage pixels. The pass or fail criteria may include discarding a PCD bearing assembly that presents such a ultrasonic testing result. The marginal band is an interval below threshold, such as an interval from the threshold down to a value of the threshold minus 10%.

[0099] The workflow for this alternative method remains similar to the one described earlier, with the main difference being the use of raw ultrasound data instead of reference images. Ultrasonic testing is performed on each PCD bearing assembly prior to placing it into service, and the resulting data is analyzed using the established thresholds and statistical criteria. If failure conditions are identified, the PCD bearing assembly may be discarded or repaired.

[0100] Calibration of the ultrasonic measurement equipment may be performed to minimize test-to-test variance, ensuring that similar cavity / PCD insert joints yield consistent results across different test runs. This calibration may involve adjustments to the gain and sensitivity of the ultrasonic transducer through modifications of the amplitude of the electrical signal applied to the transducer. Additionally, post-processing adjustments may be made to the ultrasonic measurements.

[0101] To begin, the gain and / or sensitivity are first set using the conventional method of taking measurements of a known "perfect" reflector (reference reflector), such as a flat-bottom hole or a back-wall reflection from a reference block with known material properties and dimensions. More specifically, obtaining a set of reference data includes using the ultrasonic testing apparatus on a reference PCD product with reference bonds of known bond quality. The reference PCD product may be a bearing assembly that has undergone destructive testing to confirm the actual bond quality of each joint, thereby providing reference bonds with verified and documented bond quality levels ranging from excellent to poor. Alternatively, the gain and / or sensitivity may be set by taking multiple measurements of a used component (e.g., a failed bearing assembly, with a missing PCD insert serving as a standardized reflector for this initial calibration, as this is similar to the back-wall of the part where the insert detached). The amplitude of the signals or measurements from this reference reflector (reference PCD product) is adjusted to a specific level to establish a baseline for subsequent measurements. The reference reflector is also referred to herein as reference PCD product, reference PCD bearing assembly, or calibration piece.

[0102] Calibrating the ultrasonic testing apparatus includes obtaining a set of reference data that includes more than 50 reference data points. The number of measurements used to set the gain and / or sensitivity (baseline) is at least 50 measurements (image pixels) across the reference PCD bearing product, ensuring statistical significance in the calibration process. In an alternative embodiment, at least 100 measurements may be used. In one more embodiment, at least 1000 measurements may be used. The calibration process may specifically include setting one of a sensitivity or a gain of the ultrasonic testing apparatus. The sensitivity setting controls the apparatus's ability to detect weak reflections from bond interfaces, while the gain setting amplifies the received signals to optimize the signal -to-noise ratio for accurate bond quality assessment.

[0103] In yet another embodiment, at least 1000 measurements may be used. These measurements include data from bonds of different PCD inserts located in different pockets in the reference PCD product. The used component or used bearing assembly is also referred to herein as the reference bearing assembly or reference PCD product. The reference bearing assembly is intended to represent the full range of bond qualities, from good quality to bad quality. The quality of the reference bearing assembly may have been confirmed by destructive testing methods. Alternatively, the reference bearing assembly may be an artificially created structure designed to include the full range of bond qualities (from good quality to bad quality).

[0104] The ultrasonic measurement results from the reference bearing assembly are used to adjust the sensitivity of the transducer and / or the ultrasonic testing apparatus to achieve meaningful ultrasonic image data (measurement data) for the bearing assembly under evaluation, also referred to herein as the sample bearing assembly or sample PCD product, with joints of unknown bond quality.

[0105] After this adjustment, the calibration piece (reference PCD product) with defined areas of indication (e.g., defects in a joint) is measured. This calibration piece has multiple known targets. For example, the calibration piece may be a failed PCD bearing assembly with failed joints between the body and PCD inserts, or a PCD bearing assembly specifically manufactured or modified to have the defined areas of interest.

[0106] Subsequently, a statistical evaluation of the multiple known targets is performed. The data considered for each target (e.g., each cavity / joint / PCD insert combination) may be that data inside a boundary determined based on analysis of the data, or may be that data inside a predetermined boundary.

[0107] When the calibration process utilizes images formed from the ultrasonic data, a monochrome filter is applied to the generated images. Pixels with an intensity greater than a predetermined threshold (e.g., 60%) are assigned a maximum intensity value. After applying the filter, the total number of pixels with the maximum intensity value in each image is counted. A gain factor is then applied to the ultrasonic equipment such that the total number of pixels with the maximum intensity value across all images for the bearing assembly (i.e., the sum of pixels with the maximum intensity value in each image) is adjusted to a predetermined value (e.g., 10% of the total pixels across the whole image have the maximum intensity value) in a subsequent data acquisition. If this adjustment is insufficient for the total number of pixels with the maximum intensity value across all images to meet the desired tolerance in the subsequent data acquisition due to limitations in the equipment's adjustment granularity, a corresponding gain adjustment to be made to the acquired data itself (i.e., a post-processing gain adjustment) is determined.

[0108] Finally, the calibration piece (reference PCD product, or reference PCD bearing assembly) is removed from the equipment, and ultrasonic scanning of the bearing assembly to be tested (sample PCD product, or sample PCD bearing assembly) may be performed as described above, with any gain to be made in post-processing being performed subsequent to data acquisition.

[0109] When the calibration process utilizes raw ultrasonic data, a threshold is applied to the data from the calibration piece. Data points above a predetermined threshold are assigned a constant value (or assigned a maximum value), while data points below the predetermined threshold are unchanged (or assigned a minimum value). After the application of this thresholding, the total number of data points above the predetermined threshold is counted, and a gain factor is applied to the ultrasonic equipment such that the total number of remaining data points (or total number of data points with the maximum value) across all targets (joints) for the bearing assembly is adjusted to a predetermined value (e g., 10% of the data points (pixels) have the maximum value) in a subsequent data acquisition. If this adjustment is insufficient for the total number of data points with the maximum value across all targets to meet the desired tolerance in the subsequent data acquisition due to limitations in the equipment's adjustment granularity, a corresponding gain adjustment to be made to the acquired data itself (i.e., a postprocessing gain adjustment) is determined.

[0110] After scanning the bearing assemblies to be tested (sample PCD bearing assemblies), regardless of calibration embodiment used, the calibration piece (reference PCD bearing assembly) is scanned again to verify the consistency of the calibration. The purpose of this verification step is to ensure that the ultrasonic equipment (ultrasonic testing apparatus 100) has not drifted significantly during the testing process.

[0111] When using images formed from the ultrasonic data, the total number of pixels with the maximum intensity value across all images for the calibration piece is compared to the corresponding value obtained during the initial calibration. If the difference between these two values exceeds a predetermined threshold, it indicates that the ultrasonic equipment may have drifted during the testing process, and the data acquired for the bearing assemblies to be tested is considered unreliable and should be discarded.

[0112] Similarly, when using raw ultrasonic data, the total number of data points remaining after thresholding (or the total number of data points at the maximum value) across all targets for the calibration piece is compared to the corresponding value obtained during the initial calibration. If the difference between these two values exceeds a predetermined threshold, the data acquired for the bearing assemblies to be tested is discarded.

[0113] In both cases, if the final verification step reveals that the calibration has drifted beyond the acceptable threshold, the testing process should be restarted. This involves re-calibrating the ultrasonic equipment (ultrasonic testing apparatus), scanning the sample PCD bearing assemblies to be tested again, and performing the final verification step to ensure the calibration remains consistent throughout the entire testing process.

[0114] Note that this final verification step may include additional detail. For example, in addition to comparing the total number of pixels with the maximum intensity (for image data) or the total number of data points remaining after thresholding (for raw ultrasonic data) across all targets, the variation between individual targets on the calibration piece may be inspected. This additional check helps to identify cases where the overall number of pixels or data points matches the initial calibration, but the distribution of these pixels or data points among the targets is inconsistent.

[0115] For image data, the number of pixels with the maximum intensity value for each individual target (joint) on the calibration piece is compared to the corresponding values obtained during the initial calibration. If the difference between these values for any individual target exceeds a predetermined threshold, it may indicate that the calibration may have drifted or that there may be issues with the ultrasonic equipment's performance, even if the total number of pixels with the maximum intensity value across all targets matches the initial calibration.

[0116] Similarly, for raw ultrasonic data, the number of data points remaining after thresholding (or the number of data points at the maximum value) for each individual target on the calibration piece is compared to the corresponding values obtained during the initial calibration. If the difference between these values for any individual target exceeds a predetermined threshold, may indicate that the calibration may have drifted or that there may be issues with the ultrasonic equipment's performance.

[0117] By inspecting the variation between individual targets, this final verification step may provide a more comprehensive assessment of the calibration.

[0118] Beyond the manual implementation described above, the methods presented herein are particularly well-suited for automation through specialized software applications. Such software can be configured to automatically perform the entire workflow: controlling the ultrasonic equipment, capturing the data, processing ultrasonic images or raw data, applying appropriate thresholds, executing statistical calculations to determine percentage levels of low- quality bond, and making pass or fail determinations based on the established criteria.

[0119] For the image-based method, analyzing the set of measurement data includes using an image analysis software. An automation software can implement image recognition algorithms to compare test images against reference images, quantifying the degree of similarity and determining the percentage of low-quality bond without human intervention. The software can segment each cavity / PCD insert combination, analyze the pixel intensity distribution, and match it against the reference image set to identify the closest percentage match automatically.

[0120] For the raw data method, the software can process the dB values from ultrasonic testing results, apply pre-determined thresholds, identify areas of concern, and calculate statistical values such as row averages and maximum percentages. The software can be programmed to flag specific joints of PCD inserts that exceed threshold values and to analyze patterns across rows to identify potential cascading failure risks.

[0121] The automation software can further incorporate the calibration procedures described earlier, ensuring that each test run is properly calibrated and that the results are comparable across different testing sessions. The software can automatically analyze the calibration piece data, determine the appropriate gain adjustments, and verify the calibration consistency throughout the testing process.

[0122] Additionally, the automation software can maintain a database of historical test results (historical data), which can be leveraged to continuously refine and improve the reference image sets, threshold values, and statistical criteria. This adaptive approach enables the testing methodology to evolve over time, incorporating new insights from field experiences and pattern recognition to enhance the accuracy of the joint quality assessment.

[0123] This automation capability extends beyond simple pass or fail determinations to enable advanced predictive analytics capabilities. As illustrated in FIG. 4, joint strength decays over time during operation, and the rate of decay is a critical factor in determining when a joint will ultimately fail. By establishing correlations between initial ultrasonic testing results and observed service lifespans of PCD bearing assemblies in the field, predictive models can be developed to estimate remaining useful life.

[0124] For example, bearing assemblies with higher percentages of low-quality bond (but still below the pass or fail threshold) may be assigned shorter predicted service intervals before inspection or replacement is recommended. Statistical analysis of field data, including the time- to-failure for PCD bearings with various initial ultrasonic test results, can be used to create a predictive model that estimates the remaining useful life based on the initial joint quality assessment. This predictive capability adds significant value to the testing methodology, as it allows for more sophisticated maintenance scheduling, risk assessment, and cost optimization in downhole drilling operations. Rather than simply categorizing bearing assemblies as acceptable or unacceptable, operators can make informed decisions based on the predicted service life in relation to planned drilling operations and maintenance schedules.

[0125] The methods described herein are equally applicable to post-repair validation of PCD bearing assemblies. When a bearing assembly is identified as having deficient joints and is subsequently repaired through re-brazing or other remediation techniques, the same ultrasonic testing methodology can be applied to verify the quality of the repaired joints before returning the bearing assembly to service.

[0126] For post-repair validation, the same reference image sets or raw data thresholds used for new bearing assemblies can be applied, ensuring consistent quality standards between new and repaired components. However, additional considerations may be relevant for repaired joints, such as potential changes to the cavity geometry resulting from the repair process. To address these considerations, supplemental reference image sets specifically for repaired joints may be developed, taking into account the unique characteristics of repaired areas.

[0127] The statistical analysis for repaired joints may also incorporate additional factors, such as the number of repair cycles a bearing assembly has undergone, as multiple repair cycles may affect the overall integrity of the component. By applying the ultrasonic testing methodology to post-repair validation, operators can ensure that repaired bearing assemblies meet the same quality standards as new components, minimizing the risk of premature failure and maximizing the return on investment in repair operations.

[0128] Also, multiple sets of measurement data may be obtained from multiple sample PCD products and these multiple sets of measurement data may be stored in a database, also referred to herein as historical data or historical datasets. As additional sample PCD products are tested, an additional set of measurement data may be obtained from each additional sample PCD product. The pass or fail criterion may then be updated based on the additional set of measurement data, incorporating lessons learned from field performance and evolving understanding of bond quality relationships. After updating the pass or fail criterion, the multiple sets of measurement data previously stored in the database may then be re-analyzed based on the updated pass or fail criterion, allowing for retrospective evaluation of previously tested bearing assemblies and continuous improvement of the testing methodology.

[0129] The updating of the pass or fail criterion may specifically include updating a threshold value used to distinguish between acceptable and unacceptable bond quality. This threshold updating may involve adjusting the decibel level threshold for raw ultrasonic data analysis or modifying the pixel intensity threshold for image-based analysis, based on correlation with field performance data and statistical analysis of the expanded dataset

[0130] Referring now to FIG. 14, there is shown an ultrasonic testing apparatus 100 for implementing the methods described herein for evaluating joint quality of poly crystalline diamond (PCD) inserts in bearing assemblies.

[0131] The apparatus 100 includes a power supply 122 providing power to all components. The apparatus 100 includes a signal generator 124 that produces electrical signals that are provided to a transducer 15. The transducer 15 is positioned at the inner diameter of the bearing assembly 10 and converts the electrical signals into outgoing ultrasonic waves 16 that are directed toward the bearing assembly 10. As described hereinabove, the wavelength of the ultrasonic waves is selected such that they can penetrate the steel body of the bearing assembly and interact with the joint, and the size of the beam is set to provide adequate resolution to enable inspection of the joint area and determination of the size / perimeter of the joint area.

[0132] The outgoing ultrasonic waves propagate through the steel body and into the joint, and are reflected back to the transducer 15 from the interfaces between different materials (e.g., steel / filler material, filler material / PCD) and from any defects or discontinuities in the joint, as explained in the detailed description above. The transducer 15 receives these reflected ultrasonic waves 116 and converts them back into electrical signals.

[0133] The electrical signals from the transducer 15 are provided to a signal measurement unit 123, which amplifies, filters, and digitizes the signals for analysis, as would be understood by one skilled in the art. The measurement unit 123 provides the conditioned signals to a processing unit 125that implements the methodologies described in previous sections for evaluating joint (bond) quality, including the comparison of test images to reference images or the analysis of raw ultrasound data as described hereinabove. In embodiments the measurement unit 123 and the processing unit 125 are compined in only on processor. The apparatus 100 includes a memory and or a database 128 for storing the set of reference images corresponding to the PCD bearing assembly type of interest, as well as software for implementing the statistical calculations described earlier (image analytics software), such as determining the percentage of low-quality bond in each joint and identifying the presence of failure conditions based on the established pass / or fail criterion / criteria.

[0134] The results of the analysis performed by the processing unit 125 are presented on a display 127, which may show visual representations of the ultrasonic data, such as the images depicted in FIGS. 6, 8, and 9, as well as statistical analyses as shown in FIG. 12 and information about and on a pass or fail decision. The results may also be stored in the memory and / or data base.

[0135] The apparatus 100 further provides calibration data 107 to the measurement unit 123 and / or the processing unit 125. The availability of the calibration data 107 enables the implementation of the calibration procedures described in detail above, including the setting of gain and sensitivity, the use of calibration pieces (reference PCD products) with defined areas of indication, and the verification of calibration consistency throughout the testing process.

[0136] In operation, as described previously, the transducer 15 may be mounted on a fixed head of a rotating body (FIG. 5) that can be inserted into the bore of the bearing assembly, or alternatively, a rotating head mounted on a fixed body may also be utilized. This allows the transducer to scan the entire circumference of the bearing assembly as it moves along the axial direction, ensuring full coverage of all the PCD inserts.

[0137] As detailed above, a couplant, such as a specialized fluid, is applied between the transducer 15 and the inner diameter surface of the bearing assembly 10 to facilitate the transmission of the ultrasonic waves 16 into the material.

[0138] The apparatus 100 enables the implementation of the ultrasonic testing methods described in this disclosure with sufficient precision and reproducibility to effectively evaluate joint (bond) quality in PCD bearing assemblies, thereby preventing the premature failures and cascading failure modes detailed earlier in this disclosure.

[0139] It is evident that modifications and variations can be made to what has been described and illustrated herein without departing from the scope of this disclosure. Indeed, while the present disclosure has been described in the context of evaluating the joint quality of polycrystalline diamond (PCD) inserts in a bearing assembly, the principles and techniques discussed herein can be applied to a wide range of insert types and assembly configurations. The methods presented, including the generation of reference images, statistical analysis to establish pass or fail criteria, ultrasonic testing, and comparison of test images to reference images, can be adapted to assess the bond quality of various types of inserts in different assemblies.

[0140] For example, the techniques described could be used to evaluate the joint quality of inserts made from materials other than PCD, such as cubic boron nitride (CBN), tungsten carbide, or other hard materials. Additionally, the methods can be applied to assemblies beyond bearing assemblies, to extend to any kind of assembly having cavities into which inserts are connected. Indeed, the disclosed methods can be used to non-destructively assess the joint quality and ensure the reliability and performance of a wide range of insert-based assemblies. Note that a PCD product is described here to be a bearing assembly. In one more embodiments, a PCD product may be a PCD drill bit including PCD cutters j oined or bonded in pockets of a bit body.

Claims

CLAIMS:

1. A method of evaluating a joint of a PCD insert in a PCD product, the method comprising: obtaining, using an ultrasonic testing apparatus, a set of reference data of a reference PCD product, the reference PCD product including a plurality of reference PCD inserts in a plurality of reference pockets in a reference support structure, the plurality of reference PCD inserts joined to the reference support structure by a plurality of reference bonds; calibrating the ultrasonic testing apparatus using the set of obtained reference data; obtaining, using the calibrated ultrasonic testing apparatus, a set of measurement data of a reference PCD product, the sample PCD product including a plurality of sample PCD inserts in a plurality of sample pockets in a sample support structure, the plurality of sample PCD inserts joined to the sample support structure by a plurality of sample bonds; defining a pass or fail criterion for the sample PCD product; analyzing, using a processor, the set of measurement data and applying the pass or fail criterion; and operating the sample PCD product in a wellbore when the analyzed set of measurement data meets the pass criterion.

2. The method of claim 1, wherein the set of measurement data are based on reflections of an ultrasound wave emitted by the calibrated ultrasonic testing apparatus at an interface between the sample bond and the sample support structure.

3. The method of claim 1 , wherein defining a pass or fail criterion includes defining multiple pass or fail criteria.

4. The method of claim 1, wherein optaining a set of reference data includes using the ultrasonic testing apparatus on a reference PCD product with reference bonds of known bond quality.

5. The method of claim 1, wherein analyzing the set of measurement data includes defining a threshold, and then determining whether each item of measurement data in the set of measurement data is above or below the threshold.

6. The method of claim 5, wherein each sample bond in the plurality of sample bonds is represented by a subset of measurement data of the set of measurement data; and wherein the analyzing includes determining how many measurement data in each subset of measurement data are above the threshold.

7. The method of claim 6, wherein the pass or fail criterion defines how many of the measurement data of each subset of measurement data are allowed to be above the threshold; and wherein the analyzing includes rating each of the sample bonds of the plurality of sample bonds as passed or failed based on the pass or fail criterion.

8. The method of claim 7, wherein analying the set of measurement data includes defining a marginal band, and then determining whether each item of the measurement data of the set of measurement data is within the marginal band.

9. The method of claim 8, wherein the pass or fail criterion defines how many sample bonds of the plurality of sample bonds are allowed to be within the marginal band while still maintaining operability of the sample PCD product.

10. The method of claim 1, wherein obtaining the set of measurement data includes transmitting an ultrasound wave having a first amplitude and receiving a reflected ultrasound wave having a second amplitude, determining a ratio of the first amplitude and the second amplitude, and using the ratio to analyze the set of measurement data.

11. The method of claim 1, wherein the PCD product is a bearing assembly.

12. The method of claim 1, wherein obtaining, using the calibrated ultrasonic testing apparatus, a set of measurement data of a sample PCD product comprises transmitting an ultrasound wave from a transducer, wherein the ultrasound wave first passes a portion of the sample support structure and then enters a sample bond of the plurality of sample bonds.

13. The method of claim 1, further comprising: obtaining multiple sets of measurement data from multiple sample PCD products, each sample PCD product being evaluated according to the method of claim 1; storing the multiple sets of measurement data in a database; obtaining an additional set of measurement data from an additional sample PCD product using the calibrated ultrasonic testing apparatus; updating the pass or fail criterion based on the additional set of measurement data; and re-analyzing the multiple sets of measurement data stored in the database based on the updated pass or fail criterion.

14. The method of claim 13, wherein updating the pass or fail criterion includes updating a threshold.

15. The method of claim 1, wherein analyzing the set of measurement data includes using an image analysis software.

16. The method of claim 1, wherein calibrating the ultrasonic testing apparatus includes obtaining a set of reference data that includes more than 50 reference data points.

17. The method of claim 16, wherein calibrating the ultrasonic testing apparatus includes setting one of a sensitivity or a gain of the ultrasonic testing apparatus.

18. A system for evaluating a joint of a PCD insert in a PCD product, the system comprising: an ultrasonic testing apparatus configured to obtain a set of reference data of a reference PCD product, the reference PCD product including a plurality of reference PCD inserts in aplurality of reference pockets in a reference support structure, the plurality of reference PCD inserts joined to the reference support structure by a plurality of reference bonds; wherein the ultrasonic testing apparatus is further configured to: be calibratable based upon the set of obtained reference data; and obtain a set of measurement data of a sample PCD product, the sample PCD product including a plurality of sample PCD inserts in a plurality of sample pockets in a sample support structure, the plurality of sample PCD inserts joined to the sample support structure by a plurality of sample bonds; and a processor configured to: define a pass or fail criterion for the sample PCD product; analyze the set of measurement data and apply the pass or fail criterion; and operating the sample PCD product in a wellbore when the analyzed set of measurement data meets the pass criterion.

19. The system of claim 18, wherein the set of measurement data are based on reflections of an ultrasound wave emitted by the calibrated ultrasonic testing apparatus at an interface between a given sample bond of the plurality of sample bonds and the sample support structure.

20. The system of claim 18, wherein the processor is configured to define multiple pass or fail criteria.

21. The system of claim 18, wherein the ultrasonic testing apparatus is configured to obtain the set of reference data using a reference PCD product with reference bonds of known bond quality.

22. The system of claim 18, wherein the processor is configured to analyze the set of measurement data by defining a threshold, and then determining whether each item of measurement data in the set of measurement data is above or below the threshold.

23. The system of claim 22, wherein each sample bond in the plurality of sample bonds is represented by a subset of measurement data of the set of measurement data; and wherein the processor is configured to determine how many measurement data in each subset of measurement data are above the threshold.

24. The system of claim 23, wherein analying the set of measurement data includes defining a marginal band, and then determining whether each item of the measurement data of the set of measurement data is within the marginal band.

25. The system of claim 24, wherein the pass or fail criterion further defines how many of the sample bonds of the plurality of the sample bonds in the PCD product are allowed to be rated as failed while still maintaining operability of the sample PCD product.

26. The system of claim 25, wherein the pass or fail criterion defines how many sample bonds of the plurality of sample bonds are allowed to be within the marginal band while still maintaining operability of the sample PCD product.

27. The system of claim 18, wherein the ultrasonic testing apparatus is configured to obtain the set of measurement data by transmitting an ultrasound wave having a first amplitude and receiving a reflected ultrasound wave having a second amplitude, determining a ratio of the first amplitude and the second amplitude, and using the ratio to analyze the set of measurement data.

28. The system of claim 18, wherein the PCD product is a bearing assembly.

29. The system of claim 18, wherein the ultrasonic testing apparatus includes a transducer configured to transmit an ultrasound wave, wherein the ultrasound wave first passes a portion of the sample support structure and then enters a sample bond of the plurality of sample bonds.

30. The system of claim 18, wherein the ultrasonic testing apparatus is configured to obtain multiple sets of measurement data from multiple sample PCD products, the system further comprising a database configured to store the multiple sets of measurement data, wherein the ultrasonic testing apparatus is further configured to obtain an additional set of measurement data from an additional sample PCD product, and wherein the processor is configured to update the pass or fail criterion based on the additional set of measurement data and re-analyze the multiple sets of measurement data in the database based on the updated pass or fail criterion.

31. The system of claim 30, wherein the processor is configured to update the pass or fail criterion by updating a threshold.

32. The system of claim 18, wherein the processor includes an image analysis software configured to analyze the set of measurement data.

33. The system of claim 18, wherein the ultrasonic testing apparatus is configured to be calibrated using a set of reference data that includes more than 50 reference data points.

34. The system of claim 33, wherein the ultrasonic testing apparatus is configured to be calibrated by setting one of a sensitivity or a gain of the ultrasonic testing apparatus.

Images