System and Method for Monitoring Rod Rotation and Position Using RF Signal Characteristics

Abstract

Methods and systems for monitoring rotation and position of a rod-string in a pumpjack by monitoring RF signal characteristics of wireless transmissions. A wireless device, such as a wireless load cell, is positioned to move relative to a stationary receiver during pumpjack operation. As the wireless device moves (through rotation above a rod rotator and/or linear motion during pumping strokes), RF signal characteristics vary predictably. Monitored characteristics may include signal strength, link quality, attenuation, RSSI, SNR, bit error rate, packet error rate, packet loss rate, communication interruption rate, and other RF parameters. The rod-string acts as an RF obstruction, creating periodic variations when the wireless device rotates. Rotational status, rate, and direction are determined from variation patterns. Position within the stroke cycle may also be determined. Rotator failure is detected through absence of expected variation patterns. The invention enables rotation and position monitoring without dedicated sensors, using existing wireless equipment.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

This application is a divisional of U.S. application Ser. No. 18/513,548, filed Nov. 18, 2023, which claims priority to U.S. Provisional Application No. 63/384,308, filed Nov. 18, 2022, U.S. Provisional Application No. 63/479,728, filed Jan. 12, 2023, and U.S. Provisional Application No. 63/579,003, filed Aug. 26, 2023, all of which are incorporated herein by reference in their entirety.

This application is also related to U.S. application Ser. No. 16/228,233, entitled “Apparatus and Method for Detecting the Rotation of a Rod-String in a Wellbore,” which claims priority to U.S. Provisional Application No. 62/612,503, filed Dec. 31, 2017. The disclosures of U.S. Ser. No. 16/228,233 and 62/612,503 regarding the general monitoring of RF signal characteristics for rotation detection are incorporated herein by reference.

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

Not applicable.

THE NAMES OF THE PARTIES TO A JOINT RESEARCH AGREEMENT

Not applicable.

FIELD OF THE INVENTION

The present disclosure relates to systems and methods for monitoring a rod-string in a pumpjack system. More particularly, the disclosure relates to detecting rotation and position of a rod-string by monitoring radio frequency (RF) signal characteristics of wireless transmissions from a wireless load cell or other RF transmitter. Further, concepts for integrating rotation sensing along with load cell and/or position sensing are disclosed herein.

BACKGROUND OF THE INVENTION

This section introduces aspects of the art that may be associated with exemplary embodiments of the present disclosure. This discussion is believed to assist in providing a framework to facilitate understanding of particular aspects of the present disclosure. Accordingly, this section should be read in this light, and not necessarily as admissions of prior art.

A rod pumped oil well utilizes a long slender bar, or rod, extending down into a wellbore, through production tubing, to actuate a downhole pump. The downhole pump pushes fluid up to the surface through production tubing. The wellbore path is not always perfectly vertical. Wellbore deviations, whether intentional or unintentional, cause sideload and frictional wear as the rod-string moves within the production tubing.

In order to evenly distribute this frictional wear, the rod-string is rotated at surface by a slow geared rotation mechanism known as a rod rotator. The rod rotator mechanism is subject to mechanical failure. Due to the slow rotation rate, it may be difficult to visually observe a failed rotator. Failed rotation may go undiagnosed for an extended period, resulting in uneven wear and eventual downhole failure.

Additionally, accurate position sensing of the rod-string during its cyclical pumping motion is necessary for optimal pumpjack control and diagnostics. Conventional approaches require dedicated sensors with specific placement requirements and alignment constraints.

There exists a need for a method to detect both rotation and position of a rod-string using existing wireless load cell equipment without requiring additional dedicated sensors.

TECHNOLOGY IN THE FIELD OF THE INVENTION

Radio transmitters are commonly used in the field of the invention. The “signal strength” of a transmission link is generally assessed for “good” or “bad” and the antennas adjusted such that the signal is “good”. This loose interpretation of signal quality is sufficient for both common and advanced use cases of permanently installed load cell equipment. Monitoring the signal itself, beyond the basic stability of the link, is generally not performed or necessary. Once a transmission link is established, only errors in the transmission cause a reassessment of the link quality.

Load cells and specifically wireless load cells are commonly applied in the field of the invention. The purpose of the load cell is to measure well performance. The wireless connection is used for convenience since the pumping system is constantly moving and wired connections are subject to motion-induced failure. The wireless link is a means to an end (eliminating the cable), as such, the link itself has not been utilized as an active sensing element for anything but the common link quality for the sake of stable communication.

Industry practice for wireless load cell installation specifically teaches positioning the transmitter antenna to face the receiver for optimal link quality, and placing the load cell below the rotator where it remains rotationally stationary. Manufacturers'installation guides universally recommend maximizing signal strength and minimizing obstructions. Wireless load cell manufacturers have sold thousands of units over the past decade. Despite the RF characteristics being continuously available at every receiver, no manufacturer or operator recognized that these characteristics could indicate rotation status.

There are two primary reasons why this has not been considered or developed in the industry. First, the common placement of the wired load cell is rotationally stationary, under the rotator mechanism. Wireless load cells can be installed above the rotator where they may rotate, but there appeared to be little motivation to consider alternate placement. Likewise, there appeared to be no benefit in doing so, only risk to the stability of communication. Second, the concern with stable communication necessitated optimal antenna alignment in some wireless load cells. With the advent of improved RF transmitters, it is possible to allow the transmitter to rotate and ensure stable communication.

Industry participants instead developed and sold dedicated rotation sensors as separate products, evidencing that the rotation-sensing capability of existing wireless links was not recognized or considered obvious.

Rotation Detection Using Receive Signal Strength

The general concept of Radio Signal Strength monitoring to determine rotation, presented in U.S. Ser. No. 16/228,233, is also applicable to an existing system, namely a wireless load cell, specifically in the associated receiver. The wireless load cell provides a convenient and pre-existing transmitter and can be positioned such that it undergoes rotation along with the rod-string. The receiver can then be implemented to observe the signal strength and determine if rotation is in fact occurring at the point of transmission as the transmitter is periodically obscured by the polished rod and changing direction of the transmitting antenna. Antennas have a distinct radiation pattern, and this pattern can be significantly altered when placed near a steel bar, as is the case with the installation location of the wireless load cell on the polished rod. This radiation pattern can then be utilized advantageously by observing stronger and weaker signals during the course of stroking and/or during the course of rotation.

In configurations where a wireless data connection is used for conveying data from the rotating system to a remote stationary receiver, the antenna may take several forms: an internal PCB antenna; an internal antenna but separate from the circuit board; or a distinct external antenna. The transmitter may be any radio that is rotationally bound to the polished rod and is not necessarily a dedicated sensing device for distinctly and directly monitoring rotation. A wireless load cell mounted in a manner in which its transmitter is rotating with the polished rod would achieve the goal of providing a monitorable signal of cyclically increasing and decreasing strength. In fact, any transmitter affixed to a rotating member would provide a signal which could be monitored for the cyclically increasing and decreasing strength indicating rotation.

BRIEF SUMMARY OF THE INVENTION

The present invention provides methods and systems for detecting rotation and position of a rod-string in a pumpjack by monitoring radio frequency (RF) signal characteristics of wireless transmissions from a wireless load cell device or other RF transmitter.

The RF signal characteristics that may be monitored include, but are not limited to: signal strength, signal power, signal amplitude, signal quality, link quality, communication quality, attenuation, reception strength, reception quality, received signal strength indication (RSSI), signal-to-noise ratio (SNR), carrier-to-noise ratio (CNR), bit error rate (BER), packet error rate (PER), frame error rate (FER), packet loss rate, packet drop rate, dropped packet count, communication interruption rate, link margin, fade margin, path loss, transmission success rate, retransmission rate, and connection stability.

While this disclosure frequently references wireless load cell systems as the preferred embodiment, the invention is not limited to wireless load cells. The methods and systems disclosed herein are applicable to any wireless RF transmitter positioned such that it undergoes relative motion with respect to a remote receiver, wherein the RF transmission characteristics vary with the position or orientation of the transmitter. Wireless load cells represent an already-installed system in many pumpjack applications, and the methods disclosed herein demonstrate how to utilize this existing infrastructure in a novel way to achieve rotation and position detection without requiring additional dedicated sensors.

According to one aspect of the present disclosure, a method comprises: positioning a wireless load cell device such that it moves relative to a stationary receiver during pumpjack operation; transmitting RF signals from the wireless load cell to the receiver; monitoring one or more RF signal characteristics at the receiver, wherein the RF signal characteristics vary as the load cell moves relative to the receiver; detecting variation patterns in the RF signal characteristics; and determining operational status of the rod-string based on the variation patterns.

According to another aspect, the wireless load cell is positioned above a rod rotator such that it rotates with the rod-string. As the load cell rotates, the rod-string acts as an RF obstruction between the wireless load cell antenna and the stationary receiver, creating a detectable variation in RF transmission characteristics. The RF characteristics tend to follow a periodic pattern, such as a sinusoidal waveform, with extrema occurring at specific rotational positions corresponding to the most and least optimal RF signal paths.

According to a further aspect, the RF signal characteristics also vary as the load cell moves linearly during the pumpjack stroke. The changing RF transmission path throughout the stroke creates variations in attenuation, reception strength, and other RF parameters that can be used to detect rod-string position within the stroke cycle.

According to yet another aspect, both rotation and linear position may be determined simultaneously by analyzing different aspects of the RF signal characteristic variations. Rotational status is indicated by periodic patterns with a period matching the rotational cycle, while linear position is indicated by variations correlated with the stroke cycle.

The invention provides significant advantages including: no additional dedicated rotation or position sensor hardware required; parasitic sensing using existing wireless load cell transmissions; automatic reference generation from RF characteristic extrema; immediate detection of rotator failure through absence of expected variation patterns; and integration with existing pumpjack controllers. Importantly, the technique does not consume any additional battery life as all processing is done on the receiver, which is externally powered. No impact on battery life is a substantial benefit.

Alternatively, active sensing components may be incorporated into a wireless load cell to provide additional information regarding rotational status and/or linear position. Accelerometers are commonly incorporated into wireless load cells to calculate position through a double integration of measurements. Acceleration measurements are integrated over time to obtain velocity, and the velocity values again are integrated to obtain position. A magnetometer or a barometer can be used to supplement the acceleration derived position by providing an absolute measure of an extreme position. These include pressure at the top/bottom of stroke for a barometer, or proximity to distorted magnetic fields at the top or bottom of the stroke (horsehead at the top, and wellhead at the bottom). Processing the barometric pressure is relatively straightforward as pressure drops with elevation. The magnetometer on the other hand may require extensive processing to identify the ends of stroke as distinct from mid-stroke distortions.

Additionally, a magnetometer or a gyro may be used to determine rotational changes during the course of a stroke. These measurements and observations can indicate the presence of rotation. In some cases, the measurements can be used to identify a phenomenon of torque release caused by rotational resistances downhole. Small changes in the deflection of the bridle can be measured by an accelerometer, a gyro, or a magnetometer. In some cases, this may even be detectable by the RF signal itself.

The load cell itself may be configured to measure rotational torque in addition to the vertical compressive loads of its common case. The load cell necessarily resides along the path of rotational torque, whether it is placed below the rotator (common application), or above the rotator (as discussed herein). The nature of the existing strain gauges contained in the wireless load cell may provide a measure of torque through advanced analysis.

Alternatively, special strain gauges may be placed in an orientation such that they directly measure torque, rather than compressive loads. Direct torque measurement is beneficial to well operations by indicating potential downhole failures early in their progression, perhaps early enough to take corrective action.

While many of these require hardware changes (additional measurement channels or sensors), the RF analysis approach is achievable on existing hardware by observing the already measured RF signal at the receiver.

Placement Above the Rotator Mechanism

The polished rod load cell is traditionally placed below the rotator mechanism where it is rotationally stationary. This is because traditional load cells are wired and cannot undergo rotation without a complicated wiring rotational conductor. Wireless load cells are also traditionally installed below the rotator mechanism and are positioned such that the antenna is pointing in the direction of the receiver for optimal transmission quality and signal strength.

With improvements in RF equipment, the wireless load cell may be placed above the rotator mechanism where the load cell would undergo rotation along with the polished rod. This is not a common practice, but in some cases this placement is necessary. For example, a jack-screw (for adjusting pump spacing) integrated with the rotator mechanism may create a large diameter below the rotator. The operational nature of the load cell as a safety monitoring device requires a ‘good enough’ transmitter capable of maintaining connection while antennas are not directly facing.

Not all wireless load cells can maintain a solid transmission link with the antennas facing opposite directions. This is an RF engineering problem, not a rod-pump or rotation problem. The solution has traditionally been to get a better transmitter or align the antennas. This is why monitoring the RF signal for indication of rotation has not been considered in the art of rod pumping, rod rotation, and wireless load cells.

Placing the wireless load cell above the rotator mechanism can be done selectively on installation. There is no inherent reason to place a wireless load cell below the rotator besides traditional common practice and RF link quality. If link quality is sufficient to maintain transmission, placing the load cell above the rotator is trivial. This then creates the necessary condition for monitoring the RF signal at the receiver, namely a varying signal strength. No new hardware needs to be installed on the well, the placement of the load cell (transmitter) is the key alteration. From there, the receiver needs to be modified such that it can interpret the varying signal strength.

Sensing Rotation of the Polished Rod Through a Rotationally Stationary Load Cell

Another approach to the housing of a rotation sensor would be to integrate the magnetometer into the load cell. Alternatively, a gyro, or an accelerometer, could be used for rotation sensing and likewise integrated into the load cell. In some load cell configurations, a gyro or accelerometer may be present for the purpose of determining linear stroke position through a double integration process of acceleration to velocity and again to position.

U.S. Pat. No. 10,302,510 B2 (Baker) discloses a wireless load cell assembly. Baker does not disclose rotation sensing but does describe the polished rod is free to rotate within the load cell. Likewise, the load cell itself is free to rotate with the polished rod. A traditional wired load cell requires an anti-rotation device so that the cable will not become wrapped around the assembly. Because the Baker load cell is wireless, it does not require the anti-rotation device and may be placed above the rotator such that it rotates with the polished rod.

U.S. Pat. No. 11,319,794 B2 (Fyfe) discloses a magnetometer-based rotation sensor and a wireless load cell. Likewise, U.S. Pat. No. 11,542,938 B2 (Zhao) discloses a Hurst style switch-based rotation sensor integrated into a wireless load cell. Neither of these disclose using RF signal monitoring in conjunction with a wireless load cell. Both of these treat the load cell and rotation sensor as a combination of distinct items.

U.S. Pat. No. 11,814,948 B2 (Phillips, current inventor) discloses RF monitoring as one type of rotation sensor. Only one mention of a load cell is made and not in the context of a wireless load cell. Wireless load cells are common, but positioning the wireless load cell such that it undergoes rotation is not common practice in the industry. Leveraging the existing RF link provided by the wireless load cell for rotation sensing is novel.

Yet another approach would be to integrate the magnetometer into the load cell but position the load cell in a rotationally stationary configuration, under the rotator, as a typical placement of the load cell. In this case, the magnetometer would not be observing its orientation within the Earth's magnetic field as it is rotationally stationary with respect to that field. Rather, the magnetic field of the polished rod itself or an optionally attached magnet would provide the field sensed by the magnetometer. The magnetic field of the rod-string is highly variable from one installation to another, but there is consistently “some” degree of magnetization of the rod-string. The rod naturally extends into the earth and thus conducts some amount of magnetic field. An external magnet may be attached along the polished rod, along the bridle assembly, or at the stationary stuffing-box or wellhead to increase this magnetic field and to provide a stronger and more distinct magnetic field to detect or measure.

A load cell with an anti-rotation device is known to be stationary even when the rods are rotating. In this configuration, a magnetometer placed in the load cell would observe the rotation of the magnetic field of the polished rod. This is functionally equivalent to the disclosure in U.S. Ser. No. 16/228,233, but the frame of reference is inverted. Instead of the device changing orientation within the static magnetic field of the Earth, the magnetic field, originating from the polished rod (and an optionally attached magnetic distortion) is changing orientation relative to the statically positioned magnetometer.

BRIEF DESCRIPTION OF THE DRAWINGS

In order that the disclosure may be well understood, there will now be described various forms thereof, given by way of example, reference being made to the accompanying drawings, in which:

FIG. 1 is a schematic view of a pumpjack machine with a wireless load cell device and base receiving unit in accordance with an exemplary embodiment of the present disclosure;

FIG. 2 is a detailed view of the wireless load cell device positioned above a rod rotator on the polished rod-string of the pumpjack machine of FIG. 1;

FIG. 3 is a block diagram of the wireless load cell device illustrating the RF transceiver and optional sensors;

FIG. 4A is a graph showing RF signal strength versus time during rotation of the rod-string, illustrating the characteristic periodic variation pattern;

FIG. 4B is a graph showing rod load versus time during a pumping stroke cycle;

FIG. 4C is a graph showing both RF signal strength and rod load plotted versus time, illustrating the different periodicities of rotation-induced and stroke-induced signal variations;

FIG. 5A is a conceptual diagram showing the horizontal plane radiation pattern of the wireless load cell antenna, and the receiving antenna. In this orientation, the antennas are roughly in alignment and a peak, or optimal, signal should be achieved;

FIG. 5B shows the radiation pattern of the wireless load cell antenna as it rotates away from the receiver. The distance is unchanged, but the signal becomes sub-optimal as the antennas are no longer in alignment;

FIG. 6 is a top-down view showing the geometric relationship between the rotating load cell and the stationary receiver at four different rotational positions, illustrating how the rod obstructs the RF transmission path. Conceptually, the RF transmission path is shown and becomes sub-optimal as the orientation of the transmitter (wireless load cell) changes with respect to the receiver;

FIG. 7 is a flowchart illustrating the method for monitoring rod rotation and position using RF signal characteristics;

The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way.

DETAILED DESCRIPTION OF CERTAIN EMBODIMENTS

The following description is merely exemplary in nature and is in no way intended to limit the present disclosure or its application or uses.

Definitions

For purposes of the present application, the following definitions apply:

RF transmission refers to the electromagnetic transmission of a signal. This is primarily a digital signal, but at the physical layer it is an analog transmission of frequencies. The transport protocol, RF frequencies, or communication technology are not critical to this concept. For example, the technique disclosed herein would work similarly on a Wi-Fi signal, cellular, Bluetooth, or any proprietary technology. Long range, low power technologies would also work as the technique only depends on the link metrics, not the communication itself.

“RF signal characteristics,” “RF transmission characteristics,” “RF parameters,” “RF metrics,” or “wireless signal characteristics” refer interchangeably to any measurable parameter of a radio frequency signal that can vary based on transmission conditions. Such characteristics include, but are not limited to: signal strength, signal power, signal amplitude, signal quality, link quality, communication quality, attenuation, reception strength, reception quality, reception characteristics, received signal strength indication (RSSI), received signal strength indicator, signal-to-noise ratio (SNR), carrier-to-noise ratio (CNR), bit error rate (BER), packet error rate (PER), frame error rate (FER), packet loss rate, packet drop rate, dropped packet count, communication interruption rate, communication failure rate, link margin, fade margin, path loss, propagation loss, transmission success rate, retransmission rate, acknowledgment failure rate, timeout rate, and connection stability. These characteristics may be measured as instantaneous values, time-averaged values, statistical distributions, or counts over a period of time.

“RF transmission path,” “RF path,” “signal path,” “propagation path,” or “communication path” refers to the physical path or channel through which radio frequency electromagnetic waves travel from a transmitter to a receiver. The path characteristics are affected by distance, obstructions, reflections, diffractions, and other factors that influence signal propagation. An “obstructed path” or “obscured path” refers to a transmission path in which an object such as the polished rod is positioned between the transmitter and receiver, creating increased attenuation or signal degradation.

“Attenuation,” “signal attenuation,” “path loss,” “propagation loss,” or “signal degradation” refers to the reduction in signal strength, power, or quality as the RF signal travels from transmitter to receiver. Attenuation may be caused by distance (free space path loss), obstructions in the transmission path, absorption, scattering, or other factors affecting signal propagation.

“Receive Signal Strength Indication,” “Received Signal Strength Indication,” “Received Signal Strength Indicator,” “RSSI,” “receive signal strength,” “reception strength,” or just “signal strength” is a measure of wireless link signal strength observable at the receiving end. A transmitter sends a signal with a certain amount of power. Antenna orientation and path obstructions reduce the signal power at a distance from the transmitter. RSSI is commonly expressed in dBm (decibel-milliwatts), which is a unit of level indicating that a power level is expressed in decibels (decibels referenced to one milliwatt). RSSI or dBm is typically a negative number from approximately −127 to zero, where higher (less negative) values indicate stronger signal strength. RSSI is analogous to “signal bars” displayed on a cellular phone.

“Communication interruption,” “communication dropout,” “communication failure,” or “link interruption” refers to temporary or intermittent loss or severe degradation of the wireless communication link. “Dropped packet,” “lost packet,” or “failed transmission” refers to a data packet that fails to be successfully received or acknowledged. “Periodic communication interruption” or “periodic packet loss” refers to communication interruptions that occur at regular intervals, which may indicate rotation of the wireless load cell as the rod periodically obstructs the RF transmission path.

“Link quality,” “communication quality,” “signal quality,” or “connection quality” refers to the overall quality of a wireless communication link, which may be characterized by one or more metrics including bit error rate, symbol error rate, packet error rate, frame error rate, signal-to-noise ratio, carrier-to-noise ratio, link margin, fade margin, successful transmission rate, retransmission rate, and other measures of communication reliability and performance.

“Variation,” “variation pattern,” “signal variation,” or “temporal variation” refers to changes in RF signal characteristics over time. “Periodic variation,” “cyclic variation,” or “oscillating variation” refers to variation that repeats at regular intervals. “Pattern,” “signature,” “waveform,” or “temporal pattern” refers to the characteristic shape or sequence of signal variations over time. A “sinusoidal pattern” or “roughly sinusoidal” pattern refers to a variation that follows or approximates a sine wave function. In other words, it repeats periodically. Detection can be determined through curve analysis, or by way of a threshold. In other words, if the RSSI (or other parameter) dips below a threshold in a time window, the rotator is likely functioning. Additional thresholds or statistical analysis can be performed to improve detection. For example, a dip below a threshold, and a rise above another threshold in a time window is a more definitive detection. Time spent above and below various thresholds can further improve detection.

“Rotation,” “rotational motion,” or “angular motion” refers to movement about an axis. For purposes of this disclosure, rotation specifically refers to rotation about a generally vertical axis as the polished rod and rod-string rotate. “Rotational status” refers to information about whether rotation is occurring, the rate of rotation, or the direction of rotation. “Rotational rate” or “rotation speed” refers to how fast the member is rotating, typically expressed in revolutions per minute (RPM) or degrees per second.

“Position,” “linear position,” “vertical position,” “axial position,” “elevation,” or “stroke position” refers to the position of the rod-string in its linear (up-and-down) motion during pumpjack operation. “Stroke cycle” or “pumping cycle” refers to one complete up-and-down motion of the rod-string. “Top of stroke” and “bottom of stroke” refer to the extreme positions in the stroke cycle. “Stroke reversal” refers to the change in direction at top or bottom of stroke.

“Reference tick,” “reference point,” “position marker,” or “timing reference” refers to a signal, event, or indication that corresponds to a known position, orientation, or time within a cycle. In the context of RF signal monitoring, extrema (maximum or minimum) in RF signal characteristics can serve as inherent reference ticks without requiring dedicated external sensors. This could be a physical discrete reference as in Fyfe, or a calculated reference signal. This calculated signal may also be predicted to some extent so that it triggers in real-time based on historical analysis.

“Rotator failure,” “failure of rotation,” or “failed rotation” refers to a condition in which the rod rotator mechanism is not functioning properly, and the rod-string is not rotating as intended. This may be detected by the absence of expected periodic variation in RF signal characteristics. Or, this may be detected by a deviation from an established historic pattern.

“Parasitic sensing,” “opportunistic sensing,” or “dual-use sensing” refers to the concept of obtaining additional information (such as rotation status) from equipment (such as a wireless load cell) that is primarily installed for a different purpose (such as load measurement), without requiring dedicated additional sensors. In some respects, a single receiver may be configured to “listen” or observe many transmitters. For example, a single receiver may be used to monitor rotation on several pumping units in the vicinity simultaneously. This can be done because the signal phenomenon only requires one-way transmission, or broadcast. Any observer keyed in on that signal can observe its strength.

“Antenna radiation pattern” or “radiation pattern” refers to the directional dependence of the strength of radio waves from an antenna. Antennas may not radiate or receive equally in all directions; rather, they have preferred directions of maximum radiation or reception. The radiation pattern affects how signal strength varies as the wireless load cell antenna changes orientation relative to the stationary receiver.

“Remote receiver” refers to an RF receiver positioned near the pumping unit. For the purposes of this disclosure the receiver is located at the well site. In practical terms this may be anywhere from 5-100 feet from the transmitter, which is the range where wireless load cells generally operate. The technique disclosed herein is not practically limited by distance. Because the sensing of rotation does not require the same link quality as the transmission of data itself, the rotation detecting receiver can be placed much farther from the transmitter compared with the data receiver. The remote receiver is generally stationary, but the technique could be adapted to handle both a moving transmitter and a moving receiver. This disclosure focuses primarily on a moving transmitter and a stationary receiver.

RF Signal Characteristics Variation Principle

A wireless load cell device may be placed above a rod rotator and beneath a rod retaining clamp, thus rotating with the rod-string as the rod rotator indexes the rod angular position during each pumpjack stroke. As the load cell device rotates, the RF transmission path between the wireless load cell and a base receiving unit of the pumpjack controller changes, resulting in variation of multiple RF signal characteristics.

In the context of wireless load cell installations, the prior art teaches optimizing antenna placement for maximum signal strength and positioning load cells below the rotator to avoid rotation-induced signal degradation. The present invention contradicts this teaching by intentionally exploiting signal degradation as an information source. A sub-optimal radio link previously served no purpose and every effort was made to ensure the best possible RF connection. Sub-optimal links were tolerated only if the data connection was acceptable for operation. Intentionally allowing the signal to range in quality or strength for sensing physical motion is non-obvious in the specific application of wireless load cells, and in the vast majority of permanently installed RF equipment in general.

“Parasitic measurement,” “parasitic sensing,” “opportunistic measurement,” or “piggyback sensing” refers to obtaining measurement information from a system or signal that was designed for a different primary purpose, without modifying the primary system. In the present disclosure, rotation and position information is parasitically obtained from RF transmissions whose primary purpose is conveying load measurement data.

“Negative detection” or “absence-based detection” refers to identifying a condition (such as rotator failure) based on the absence of an expected signal or pattern, rather than the presence of a fault-indicating signal. The RF-based rotation detection employs negative detection: rotator failure may be indicated by the absence of expected periodic RF variation, not by a dedicated sensor explicitly reporting failure.

Additional RF signal characteristics that may be monitored include, but are not limited to: channel state information (CSI), received channel power indicator (RCPI), automatic gain control (AGC) level, modulation error ratio (MER), error vector magnitude (EVM), forward error correction rate, link budget margin, and any other metric derivable from RF signal reception that varies with transmission path conditions.

The RF transmission characteristics that vary with rotation include: RF energy level, link quality, attenuation, reception strength, reception quality, RSSI, and other measurable RF parameters. These characteristics tend to remain strong during the portion of rotation that faces the receiver, and tend to follow a sinusoidal or periodic waveform on the back portion of rotation with a minimum occurring when the wireless load cell RF emissions window is pointing approximately 180 degrees away from the receiver. In this orientation, the antenna is both facing away from the receiver and also obscured by the polished rod.

The mechanism compounding this variation is that the polished rod itself acts as an RF obstruction. When the wireless load cell antenna is directed toward the receiver, the RF path is relatively unobstructed. When the load cell rotates such that the polished rod is between the antenna and receiver, the RF signal must diffract around or through the rod, resulting in increased attenuation and degraded signal characteristics. This technique would work if there were no obstruction, only a directional antenna. For example, the transmitter could be put at the top of the polished rod. Ignoring external factors such as proximity to the horsehead, an omni-directional antenna placed at the top of the polished rod would not achieve the desired result, but a directional antenna would. An omni-directional antenna placed next to the polished rod, however, would achieve the desired result.

Accordingly, monitoring this variation in RF transmission characteristics may be used as a direct indicator of the rod-string's rotational position and rotational rate. The monitored RF characteristics may be reported to the pumpjack controller by the base receiving unit via analog and/or digital means. The variation signal is useful in detecting potential failure of normal rod-string rotation operation due to either failure of the rod rotator mechanism or excessive angular load on the rod-string downhole.

In addition to rotational variation, the RF transmission characteristics also vary as the wireless load cell moves linearly during the pumpjack stroke. The changing relative position between the load cell and the stationary receiver creates a changing attenuation path throughout the entire stroke. These variations in reception strength, quality, and other RF parameters throughout the stroke can be used to detect rod-string motion and position within the stroke cycle.

The rotational status is confirmed when the detected RF variation period matches the expected rotational period calculated from the rod rotator specifications and pumpjack stroke rate. For example, if the rod rotator advances 2 degrees per stroke and the pumpjack operates at 6 strokes per minute, one complete rotation requires 180 strokes (360°/2°) taking 30 minutes. Detection of RF signal variation with a 30-minute period confirms rotation. Absence of this expected periodicity, or sudden change in period, indicates rotator malfunction. The system may store rotator specifications and automatically calculate expected rotational periods for comparison with observed RF variation periods.

Specific RF Characteristics for Monitoring

RSSI (Receive Signal Strength Indication) monitoring: The receiver monitors the RSSI value of transmissions from the wireless load cell. As the load cell rotates and the rod obstructs the RF path, the RSSI decreases. The periodic variation in RSSI directly corresponds to the rotational cycle. RSSI extrema (minimum and maximum) provide inherent reference points for rotation detection.

Link quality monitoring: Modern wireless transceivers often provide link quality metrics in addition to or instead of raw RSSI values. Link quality may be expressed as a numerical score, percentage, or other metric. Monitoring link quality variation provides an alternative or complementary approach to RSSI monitoring for rotation and position detection.

Attenuation monitoring: The attenuation (signal loss) in the RF path varies predictably with load cell rotation and position. By monitoring changes in attenuation, either directly or through proxy measurements such as RSSI, the system can determine rotational and linear position.

Bit error rate or packet error rate monitoring: Communication errors increase when the RF path is obstructed. Monitoring the bit error rate or packet error rate as a function of time provides an indication of when the rod is obstructing the RF path, thus indicating rotational position.

Signal-to-noise ratio (SNR) monitoring: SNR varies with signal strength and path conditions. Monitoring SNR variation provides another mechanism for detecting rotation and position through RF characteristics.

Periodic communication interruption or packet loss monitoring: When the RF path is severely obstructed during rotation, the wireless link may experience temporary communication failures or dropped packets. By monitoring the occurrence of communication interruptions or counting dropped packets over time, periodic patterns can be detected. A periodic increase in dropped packet count or periodic communication failures occurring at intervals matching the rotational period indicates that rotation is occurring. The frequency of these periodic interruptions provides an indication of rotational rate.

This approach is particularly useful in digital communication systems where packet-level statistics are readily available. The system can count successful versus failed transmissions, monitor acknowledgment timeouts, or track retransmission requests. A periodic pattern in these metrics, synchronized with the expected rotational cycle, confirms rotation is occurring even if absolute signal strength measurements are not available or are unreliable.

In embodiments utilizing frequency-hopping or spread-spectrum communication protocols, the RF signal characteristics may be monitored across multiple frequency channels. The rotation-induced signal variations appear consistently across all channels, while frequency-specific interference affects only individual channels. By analyzing the RF characteristics across multiple frequencies, the system can more reliably distinguish rotation-induced variations from environmental noise and interference. Additionally, the differential signal strength between frequency channels may provide additional rotation-correlated information due to frequency-dependent diffraction and reflection characteristics around the polished rod obstruction.

The RF-based rotation detection is agnostic to the specific wireless protocol, frequency band, or modulation scheme employed by the wireless load cell. The rotation-induced variation in RF characteristics arises from propagation physics (obstruction, antenna directionality) that affect all RF transmissions regardless of protocol. Thus, the method applies equally to 900 MHz ISM band transmissions, 2.4 GHz transmissions, 5.8 GHz transmissions, licensed spectrum transmissions, spread-spectrum protocols, narrowband protocols, packet-based protocols, frequency-hopping protocols, and continuous-wave protocols. Specific protocols that may be employed include, but are not limited to: Bluetooth Low Energy (BLE), Zigbee, LoRa, LoRaWAN, Wi-Fi (802.11 variants), cellular (LTE, 5G NR), proprietary industrial protocols, and any other RF communication protocol.

One-way transmission is all that is required for this technique. Additional modifications could include 2-way communications where the link is monitored in both directions. For this to be practical, the wireless load cell would need to include these statistics in the data transmitted. This would add minimal additional battery consumption in the form of extra bytes transmitted.

Combined Monitoring of Multiple RF Characteristics

In some embodiments, multiple RF signal characteristics are monitored simultaneously. For example, both RSSI and link quality may be monitored to provide redundant rotation detection or to improve accuracy. Different RF characteristics may have different sensitivities to rotation versus linear motion, and combining multiple measurements can help distinguish between rotational and linear position changes.

The combination of RF characteristic monitoring with other sensor inputs (such as a magnetometer, accelerometer, gyro, barometric pressure sensor, or strain gauge data from the load cell) can further improve system accuracy and provide comprehensive monitoring of rod-string status.

Pattern Recognition and Analysis

The receiver or controller analyzes the temporal patterns in RF signal characteristics to extract information about rod rotation and/or position. For rotation detection, the system identifies periodic variations with a period matching the expected rotational cycle time. The presence of periodic variation confirms that rotation is occurring, while the absence of periodic variation indicates a rotator failure.

The characteristics of the periodic pattern provide additional information. A sinusoidal pattern with a local minimum alternating with a local maximum indicates normal rotation with the rod acting as an obstruction. The frequency of the periodic variation indicates the rotational rate. Asymmetry or unequal periods in the pattern can indicate rotational direction, as the signal characteristics may differ on the approach to versus departure from the obstruction position. Knowledge of the pumping speed and parameters can be used to correct for non-rotational changes.

The sinusoidal characteristic of the RF variation pattern enables efficient detection algorithms. Rather than requiring complex pattern matching, the receiver can apply narrowband filtering at the expected rotational frequency to isolate rotation-induced variations from broadband noise. The predictable waveform shape also enables phase-based analysis for determining rotational position within a revolution, not merely rotation rate. Signal processing techniques applicable to rotation detection include, but are not limited to: Kalman filtering, moving average filtering, exponential smoothing, Fourier transform analysis, autocorrelation, cross-correlation, matched filtering, peak detection algorithms, threshold-based detection, and adaptive filtering.

Rotational direction may be determined by analyzing asymmetry in the RF signal variation pattern. As the wireless load cell antenna approaches the obstructed orientation (rod between antenna and receiver), the signal strength decreases at a rate determined by the antenna's radiation pattern and the rod's RF shadow. As the antenna departs from the obstructed orientation, the signal strength increases. Due to antenna directivity characteristics and multipath effects, the approach and departure signatures may differ, creating an asymmetric waveform. The system may determine whether the load cell is rotating clockwise or counterclockwise relative to the receiver by analyzing this asymmetry. For example, the slope of signal decrease to the slope of signal increase can be used to indicate the signal strength change relative to the angle of the antenna and relation to the obstruction.

For linear position detection, the system correlates RF characteristic variations with the stroke cycle. Changes in RSSI or other characteristics that correlate with the known stroke period can be used to identify stroke position, particularly stroke reversal points where the direction of motion changes.

Analyzing historic patterns of RF signal characteristics can improve detection accuracy and reduce false alarms. Historical data can be leveraged to establish baseline signatures for normal rotation, to adaptively adjust detection thresholds based on observed signal strength ranges, to learn the characteristic rotational period, and to track gradual changes in signal patterns over time that may indicate developing mechanical issues.

The system may employ adaptive threshold algorithms to automatically adjust detection sensitivity based on observed signal conditions. During initial installation or calibration, the system records baseline RF signal characteristics during confirmed rotation and non-rotation periods. These baselines establish initial detection thresholds. During operation, the thresholds are continuously refined based on observed signal statistics, seasonal variations affecting RF propagation, equipment aging, and any detected changes in the RF environment. Adaptive thresholding ensures reliable rotation detection across varying environmental conditions without requiring manual recalibration. Prior knowledge of the RF environment is not required, but may be helpful.

In embodiments where the pumpjack operates intermittently, the system may exploit idle periods for calibration. Many wells are pumped on a scheduled basis with idle periods between pumping cycles. During idle periods when the rod-string is stationary, the system characterizes RF baseline without rotation-induced variation. This ‘quiet period’ baseline is used to calibrate subsequent rotation detection during pumping periods. The contrast between quiet-period and pumping-period RF characteristics provides direct measurement of rotation-induced signal variation magnitude, enabling more accurate detection thresholds. Again, this is an improvement on detection and not a requirement.

Antenna Radiation Patterns

The variation in RF signal characteristics with rotation is influenced by the radiation pattern of the wireless load cell antenna. Antennas exhibit directional gain characteristics that can be represented by a three-dimensional radiation pattern showing signal strength as a function of angle and distance.

As the wireless load cell rotates about the polished rod, the antenna rotates in the horizontal plane relative to the stationary receiver. When the antenna is oriented such that its direction of maximum gain points toward the receiver, the received signal strength is at or near its maximum value. As the antenna continues to rotate, it passes through angles of progressively poorer alignment with the receiver, and the received signal strength decreases according to the antenna's horizontal plane radiation pattern.

When the antenna has rotated approximately 180 degrees from its optimal orientation, it is pointing directly away from the receiver. At this orientation, several factors combine to minimize the received signal strength: (1) the antenna gain in the 180-degree direction is at or near its minimum; (2) the polished rod is positioned between the antenna and the receiver, physically obstructing the direct RF path; and (3) the RF signal must propagate around or through the rod obstruction, experiencing additional attenuation.

The vertical plane radiation pattern affects signal variations during the linear (up-and-down) motion of the wireless load cell during pumping strokes. As the load cell moves vertically, the elevation angle from the receiver to the load cell changes. This elevation angle change causes variations in received signal strength correlated with the stroke cycle.

The wireless load cell system sends a significant and constant stream of data packets, each of these can be assessed for RSSI or other RF characteristics. A single measurement may contain noise, but when receiving several packets per second, filtering techniques can be used to greatly improve the detection by eliminating this noise.

Various antenna types may be employed in the wireless load cell, each producing characteristic radiation patterns that affect rotation detection. Antenna types include, but are not limited to: patch antennas, dipole antennas, monopole antennas, chip antennas, PCB trace antennas, planar inverted-F antennas (PIFA), helical antennas, and ceramic chip antennas. More directional antennas (such as patch antennas) produce larger amplitude variations in RF signal characteristics as the load cell rotates, potentially improving detection signal-to-noise ratio. Less directional antennas (such as chip antennas) produce smaller amplitude variations but may provide more uniform data transmission quality. The antenna type may be selected based on the tradeoff between rotation detection sensitivity and data transmission reliability.

Torsional Energy Detection

When excessive downhole angular loads occur on the rod-string, thus keeping the lower portion of the rod-string from rotating, the rod-string begins to store torsional energy as the rod rotator continues to rotate the top of the rod-string. This torsional energy continues to increase until it ultimately exceeds the downhole torsional load, resulting in a rapid release of the torsional energy. This condition may create excessive loads on the rod rotator leading to degradation or failure.

The RF signal characteristic monitoring may not directly observe this torsional energy buildup phenomenon because the top of the rod-string (where the load cell is attached) continues to rotate even when the lower portion is stuck. However, the RF monitoring method can be combined with other sensors to detect this condition.

A three-axis accelerometer sensor can detect rapid changes in torsional rod-string motion by observing abnormal spikes in the tangential component of acceleration. When RF monitoring indicates continued rotation, but accelerometer data shows torsional release events, the system can alert the controller to take corrective actions.

A gyroscope sensor can detect abrupt changes in rotational angle or angular velocity. When torsional energy releases, the load cell experiences a rapid change in angular position or a spike in angular velocity that differs from the normal slow, steady rotation provided by the rod rotator. By monitoring the gyroscope output for such anomalous rotational events, the system can detect torsional wind-up and release.

A magnetometer sensor can determine absolute rotational orientation relative to the Earth's magnetic field. By measuring the magnetic field direction (compass heading) at the load cell location, the magnetometer provides an absolute angular reference. Torsional wind-up causes the upper portion to rotate beyond the angular position of the lower portion, creating a measurable angular deflection. By monitoring the magnetometer reading at the same point in each stroke and comparing the rotational orientation from stroke to stroke, the system can detect progressive torsional deflection. The magnetometer can also be used to determine proximity to other objects by way of field distortions and so the magnetometer can double as a position sensor.

DESCRIPTION OF SELECTED SPECIFIC EMBODIMENTS

The preferred embodiment integrates the RF characteristic monitoring into the receiving end of an existing wireless load cell system. The receiver, which already receives load data from the wireless load cell, is enhanced with firmware or software to additionally monitor one or more RF signal characteristics of the received transmissions.

In a specific implementation, the receiver monitors RSSI values reported by its wireless transceiver. Modern wireless transceivers typically provide RSSI information as part of their standard functionality, so no additional hardware is required. The receiver firmware processes the RSSI time series to detect periodic variations indicating rotation.

When periodic RSSI variation matching the expected rotational period is detected, rotation is confirmed and the rotational rate can be calculated. When RSSI variation is absent for a predetermined number of strokes (for example, approximately 100 strokes or more), a rotator failure alert is generated and transmitted to the pumpjack controller or operator interface. Alternatively, a time threshold could be used.

The rotational status information may be communicated to the pumpjack controller via existing communication interfaces. The controller can use this information for diagnostics, maintenance alerts, or automatic control decisions such as shutting down the system when a rotator failure is detected. The field operator may also use these alerts to alter pumping system operations.

In the preferred embodiment, RF characteristic monitoring is performed primarily at the stationary receiver rather than at the transmitter. This approach is particularly advantageous for battery-powered wireless transmitters such as wireless load cells, where power consumption is a critical constraint. By performing all signal analysis at the receiver, no additional hardware, processing, or power consumption is required at the transmitter. The transmitter continues its primary function of transmitting load data without modification, while the receiver performs the additional task of monitoring RF characteristics for rotation and position detection. This approach exemplifies parasitic sensing, wherein rotation and position information is obtained without imposing any burden on the transmitter.

Performing all RF signal analysis at the stationary receiver rather than the battery-powered transmitter provides significant advantages. The receiver has access to continuous external power, enabling computationally intensive signal processing without battery constraints. The receiver can store extensive historical RF characteristic data for trend analysis, pattern learning, and anomaly detection. Receiver firmware can be updated remotely over network connections without physical access to the well site. Most importantly, the transmitter, which is often the most difficult and expensive component to access for maintenance, requires no modification whatsoever. The transmitter continues its primary function of transmitting load data without any awareness that rotation detection is being performed from its transmissions.

Alternative Embodiments and Variations

While RSSI monitoring is described as the preferred embodiment, other RF characteristics can be used alternatively or additionally. Link quality metrics, when available from the wireless transceiver, may be monitored instead of or in combination with RSSI. Bit error rate or packet loss rate monitoring provides another approach, particularly in digital communication systems where these metrics are readily available.

The wireless load cell may alternatively be positioned at other locations where it moves relative to a fixed receiver. While placement above the rod rotator is preferred for rotation detection, other configurations are possible depending on the specific monitoring goals.

The RF characteristic monitoring may be implemented entirely at the receiver, entirely at the load cell (with the load cell monitoring its own transmit and receive characteristics if bidirectional communication is available), or distributed between the load cell and receiver with coordination between the two devices.

The antenna used in the wireless load cell may be selected or designed based on its radiation pattern characteristics to optimize rotation detection performance. A more directional antenna will produce larger amplitude variations in RF signal characteristics as it rotates, potentially improving detection reliability and signal-to-noise ratio.

Other sensors such as magnetometers, accelerometers, or gyroscopes may be used in conjunction with RF monitoring to provide comprehensive rotation and position sensing. However, the key advantage of the RF monitoring approach is that it requires no additional rotation-specific sensors beyond the wireless load cell that is already present for load measurement.

The RF-based rotation and position detection method is particularly advantageous for retrofit installations on existing pumpjack systems. Unlike dedicated rotation sensors that require physical attachment to rotating components and potential modification of the bridle assembly, the RF monitoring approach requires only: (a) repositioning the existing wireless load cell above the rod rotator, which may be accomplished during routine maintenance without specialized tools or skills; and (b) updating receiver firmware to analyze RF signal characteristics, which may be performed remotely in many installations. The transmitter firmware and hardware remain completely unchanged. This enables rotation monitoring to be added to thousands of existing wireless load cell installations through software update alone, with minimal or no site visits required.

The RF-based approach provides significant cost advantages over dedicated rotation sensors. Dedicated rotation sensors require: purchase of sensor hardware, installation labor, potential modification of bridle assembly, wiring or wireless integration with existing control systems, ongoing battery replacement for wireless sensors, and periodic calibration. The RF-based approach requires: no new hardware (uses existing wireless load cell), minimal installation labor (repositioning existing equipment), no bridle modification, no additional system integration (uses existing receiver), no additional batteries, and automatic calibration through adaptive algorithms. For operators with large numbers of wells equipped with wireless load cells, the RF-based approach enables fleet-wide rotation monitoring at a fraction of the cost of dedicated sensor deployment.

Alerting mechanisms for detected rotator failure may include SCADA system alarms, text messages to designated personnel, email notifications, push notifications to mobile applications, audible alarms, visual indicators, automatic shutdown commands to the pumpjack controller, logged events in a data historian, and activation of failsafe operating modes.

Advanced Embodiments and Future Developments

The following section describes advanced embodiments that extend beyond the basic RF signal characteristic monitoring described above. These embodiments are disclosed for completeness and to provide context for potential future implementations. While certain independent claims presented herein may not encompass all aspects of these advanced embodiments, the advanced embodiments remain within the scope of this disclosure and may be the subject of additional claims in this application or continuation applications.

Time of Flight Measurements

In advanced embodiments, Time of Flight (ToF) measurements may be utilized in addition to or instead of signal strength and quality metrics. Time of Flight refers to the time it takes for an RF signal to travel from the transmitter to the receiver. By measuring the ToF, the system can determine the distance between the transmitter and receiver.

This distance varies as the transmitter moves vertically during the stroke cycle and changes subtly as the transmitter rotates. During rotation, the effective path length changes as the direct line-of-sight path is obstructed and the signal must travel via reflected or diffracted paths, resulting in measurable ToF variations.

ToF measurements require precise timing capabilities in both the transmitter and receiver, typically achieved through ultra-wideband (UWB) radio technology or other precision timing protocols. Modern UWB tracking systems, such as those used in consumer device location (e.g., Apple AirTags), demonstrate the feasibility of centimeter-level distance measurement using ToF techniques.

In the context of rod rotation and position sensing, ToF measurements can provide an additional mechanism for detecting both the vertical position (through absolute distance measurement) and rotational status (through path length variations as the rod obstructs the direct path).

Angle of Arrival Measurements

Angle of Arrival (AoA) refers to the angle at which an RF signal arrives at the receiver. For receivers equipped with multiple antennas or antenna arrays, the phase differences in the arriving signal can be used to determine the direction from which the signal originated.

As the transmitter rotates, the AoA changes, providing another mechanism for rotation detection. As the transmitter moves vertically during the stroke, the elevation angle changes, providing position information. AoA measurements can complement ToF measurements to provide more comprehensive position and orientation information.

AoA measurement requires a receiver with multiple antenna elements separated by a known distance, typically a fraction of the RF wavelength. The receiver processes the phase differences between antenna elements to calculate the arrival angle. This technique is well-established in direction-finding applications and is increasingly available in commercial wireless chipsets.

Multi-Receiver Triangulation Systems

In advanced embodiments, multiple receivers positioned at different locations may be used to provide enhanced rotation and position detection through triangulation or multilateration techniques. Each receiver independently monitors the RF characteristics of the transmitter's signals, and the collective measurements from all receivers are combined to more accurately determine the transmitter's position and orientation.

A multi-receiver approach can leverage Time of Flight (ToF) differences between receivers to triangulate the transmitter's three-dimensional position. By comparing the arrival times at multiple receivers with known positions, the system can calculate the transmitter location using geometric relationships.

Similarly, Angle of Arrival (AoA) measurements from multiple receivers can be combined to determine the transmitter's orientation relative to each receiver, providing redundant rotation detection. Comparative signal strength measurements from multiple receivers can also better characterize rotation and position by eliminating ambiguities that may exist with a single receiver.

The redundancy provided by multiple receivers improves system reliability, as rotation and position can still be determined even if one receiver experiences temporary signal loss or interference. The geometric diversity of receiver placements (for example, receivers positioned at different horizontal and vertical positions relative to the pumpjack) provides complementary information that enhances overall detection accuracy. In effect, the use of multiple receivers would be similar to GPS except that in GPS, multiple transmitters are used with a single receiver (there can be many independent receivers).

Passive Monitoring Receivers

In certain embodiments, a passive monitoring receiver may be employed. This passive receiver monitors the RF transmissions between the wireless load cell (or other transmitter) and the primary receiver without being an active participant in the communication link. In other words, the passive receiver listens to or eavesdrops on the wireless transmissions without itself transmitting or acknowledging packets.

This approach is advantageous in scenarios where adding RF monitoring capability to an existing system would be impractical, or where regulatory or protocol constraints prevent modification of the primary communication link. This is “receive only,” and simplifies development and ongoing hardware support and maintenance. The passive receiver can be added to existing installations without requiring changes to the wireless load cell firmware or the primary receiver.

The passive receiver analyzes the RF signal characteristics it observes (RSSI, packet timing, signal quality, etc.) to independently determine rotation and position status. Because the passive receiver does not interfere with the primary communication link, it provides a non-intrusive monitoring solution.

The primary limitation of passive monitoring is that certain metrics that require bidirectional handshaking (such as acknowledged packet statistics from the transmitter's perspective) are not directly observable. However, many useful metrics such as received signal strength and packet reception timing remain available to the passive receiver.

The passive monitoring receiver enables rotation detection on third-party wireless load cell systems without requiring access to, or cooperation from, the load cell manufacturer or primary system operator. An operator can deploy a passive receiver to monitor rotation status on wells equipped with existing third-party wireless load cells, using only the RF emissions that the load cell already produces. This is particularly valuable in asset acquisitions where legacy equipment from multiple vendors is present, or in situations where the primary wireless system vendor does not offer rotation monitoring capability. The passive receiver requires no modification to existing equipment, no access to proprietary communication protocols, and no integration with the primary receiver. Only physical placement is required within RF range of the varying wireless load cell transmissions.

Multipath Propagation and Echo Analysis

The signal path between transmitter and receiver may include direct line-of-sight propagation as well as echoes, scattering, and reflections from surrounding structures such as the pumpjack frame, motor housing, wellhead equipment, and ground surface. While these multipath effects can complicate simple signal strength measurements, they can also be leveraged to improve detection accuracy in advanced implementations.

Multipath propagation occurs when RF signals reach the receiver via multiple paths due to reflections, scattering, and diffractions from surrounding objects. An echo or reflected signal is a delayed copy of the transmitted signal that arrives at the receiver after bouncing off a reflective surface. Scattering refers to the redirection of RF energy in multiple directions when encountering rough surfaces or objects comparable in size to the wavelength.

By analyzing the time delays of various echo paths, similar to techniques used in radar systems, an advanced implementation can better characterize the transmitter's position and orientation. The combination of direct and reflected signal paths creates a unique RF signature for each position and orientation, potentially enhancing the ability to detect rotation and position changes.

While a simple obstruction creates a single minimum in signal strength, the combination of direct and multipath signals creates a more complex but information-rich signal pattern. Advanced signal processing techniques, including matched filtering and channel estimation algorithms, can extract position and orientation information from multipath-affected signals.

These complex signal analysis techniques are analogous to methods used in modern ultra-wide band (UWB) tracking systems and indoor positioning systems, which routinely deal with multipath-rich environments. Applying similar techniques to pumpjack monitoring could provide enhanced accuracy and robustness, particularly in installations with significant metallic structures that create strong reflections.

Environmental factors including temperature, humidity, precipitation, and atmospheric pressure can affect RF propagation characteristics independent of rod rotation or position. The system may incorporate environmental compensation algorithms that adjust for these effects. Temperature sensors within the receiver or external weather data can trigger compensation algorithms when significant environmental changes occur. In one embodiment, the system monitors the long-term baseline of RF signal characteristics (over periods of hours to days) and automatically compensates for gradual environmental drift, ensuring that rotation and position detection remain accurate throughout seasonal changes. By separating the slow timescale of environmental variation from the faster timescale of rotation-induced variation, the system distinguishes environmental effects from mechanical motion without requiring dedicated environmental sensors.

In advanced embodiments, machine learning algorithms may be applied to RF signal characteristic analysis for rotation and position detection. Supervised learning approaches may be trained on labeled data from wells with confirmed rotation status to develop classifiers that distinguish rotating from non-rotating conditions. Unsupervised learning approaches may identify anomalous RF patterns that indicate developing mechanical problems. Time-series prediction models may forecast expected RF variation patterns based on historical data, with deviations from predictions triggering alerts. Neural network architectures suitable for this application include recurrent neural networks (RNNs), long short-term memory (LSTM) networks, convolutional neural networks (CNNs) applied to time-frequency representations of RF data, and transformer architectures for sequence modeling.

Detailed Description of Figures

FIG. 1 shows a typical pumping unit. The rod rotator and bridle assembly is at 120. This includes, but is not detailed, the wireless load cell. An illustrative transmitter corresponding to the wireless load cell is shown at 160. The well automation system, pump-off controller, is shown at 150, with an illustrative antenna corresponding to the receiver of the wireless load cell at 151. The RF path does not travel in a simple line, but the illustrative path is shown at 170, linking the transmitter and receiver. Again, RF signals do not behave as illustrated here, but the illustration is provided for a conceptual understanding. When the transmitter antenna is facing away, the RF path is indirect and must “turn” at 171 to reach the receiver (again, this is figurative, not literal).

FIG. 2 shows a zoomed in view of the bridle assembly, 120. The load cell is shown at 210 and the associated transmitter is shown at 220. A typical installation would place the load cell (210) directly on the bridle 203, and under the rotator mechanism 202. When placed below the rotator at 201, the load cell would be rotationally stationary. In the preferred embodiment, the load cell is placed, as illustrated, above the rotator 202, where it undergoes rotation. This results in the transmitter 220 rotating. This means the transmitter antenna will rotate, or point in different directions while rotating.

FIG. 3 shows block components of the transmitter and receiver. The minimum requirement is the RF transmitter 220, and the preferred embodiment includes the load cell 210. Several optional sensors may be included on the transmitter. In one embodiment, as disclosed in U.S. Ser. No. 18/513,548, an optional magnet may be placed opposite a magnetometer to determine rotation. The scope of this disclosure focuses on the RF analysis to determine rotational status. The receiver 320 includes at the minimum an RF receiver 321 and a processor 322. The processor may include data storage for historical analysis and pattern matching. At a minimum, a threshold may be preconfigured to establish rotational status, or the system may be configured to determine this through a calibration process. Additionally, the system would include a status output 323 to provide connectivity or alerting to external systems.

With regards to torsional changes in the rod-string, the placement of the sensing device with the optional sensors determines how the torsional changes would be identified. A magnetometer for example may be placed on the rotationally bound bridle to determine changes in deflection of the bridle. In terms of a wireless load cell, this could be incorporated into an existing wireless load cell positioned under the rotator in the typical configuration. Alternatively, this could be done on the rotating member with some additional processing to identify net positive changes relative to sudden torsional relaxation. A gyro similarly could be placed in either the rotationally bound configuration, or in the rotating configuration. The difference between a magnetometer and a gyro is that a magnetometer can be rotationally bound and still determine rotation of a nearby member. This can be done by ensuring the magnetic field of the nearby member is sufficiently detectable in the rotational direction (i.e. attach a magnet if the inherent magnetic field is insufficient). An accelerometer could be used in addition to, or in place of, a gyro.

FIG. 4 shows several plots of the system measurements. FIG. 4A is an approximation of RSSI (signal strength) on the Y axis, and time on the X axis. When the antennas are facing each other the signal is strong, perhaps even maxed out. As the rotating antenna moves away, the signal gets weaker, shown by a dip in the signal strength. In the absence of operational knowledge, an intentionally stopped pump may cause a false alarm. A sufficient window of analysis should be used to trigger a notification. In general, this is a known quantity based on the strokes per minute and the number of rotator handle pulls per rotation. In other words, it may take 175 strokes per rotation and at 6 strokes per minute, that would take approximately 30 minutes. Note: this 30-minute period is only active during pump operation and may be discontinuous (it may take several hours to achieve the 175-stroke/30-minute rotation, if the pump frequently starts/stops).

A window of an hour may be used to trigger an alert of a rotation failure. Conversely, a more instantaneous alert could be provided by watching shorter term rotational stats during a single stroke, or a partial stroke. Rotational status is not a high priority error and it may take several hours or days to deploy a technician to investigate. In general, it is better to avoid false positives at the expense of delaying the alert. There is no technical reason the alert could not be generated immediately on detection of non-rotation during periods of pump operation.

FIG. 4B shows a typical load vs time plot. Changing load indicates the pumping system is operational. This can be used to determine periods of expected rotation. It is convenient on a wireless load cell as the RSSI (to determine rotation) and the load variations (to detect operation) are both readily available at the receiver and can be processed without additional sensors or signals. In other words, everything the system needs to determine positive rotational failures is already present in existing wireless load cell systems.

FIG. 4C overlays these 2 (load variations and RSSI variations). Of note, significantly more strokes per rotation are common in practice. For illustrative purposes, an unnaturally fast rotator is shown to make the plots clear. If the load plot goes flat, no rotation is expected. This is an expected, or non-error condition. If the load plot is cycling and the RSSI plot goes flat, the rods are not rotating during a time of expected rotation. This is an error condition. If both go flat, this is nonoperation, and not a failure state. The illustrated plot shows an operational and rotational state.

FIG. 5A shows illustrative radiation patterns for a typical directional antenna. This is a top-down view. Several “lobes” are shown, but the dominant one is shown by 530. Radiation patterns are three dimensional, but are typically drawn in the 2D horizontal and vertical planes. This represents the optimal signal envelope. The stationary receiver is shown in 520 and 521. In FIG. 5B, the rotating transmitter is shown in the optimal rotational orientation in 510, and in a sub-optimal orientation at 90 degrees in 511. This rotation continues in a clockwise motion at the transmitter mounted on the rotating member (i.e. polished rod and rod rotator table). 540 shows the location of the antenna within the radiation pattern diagram. Antennas are located at this point in the other diagrams.

FIG. 6 shows a top down view of the transmitter as it rotates clockwise 603 around the polished rod 601 (not to scale). The RF receiver is stationary and illustrated by 602. 4 rotational positions are shown for the transmitter (only one transmitter is present, but is illustrated in 4 orientations). As discussed previously, RF transmission path shown by the dashed lines is illustrative, not literal. The receiver sees a strong signal when the transmitter is facing the receiver at position 610. As the transmitter rotates away (620) from the receiver, it sees a reduction in signal strength. This is illustrated by the indirect dashed line 621. In reality, the signal is just weaker as it is on the periphery of the radiation pattern. At position 630, the transmitter is pointing directly away from the receiver, and the polished rod is in the way. This is likely to result in the weakest signal, but reflections in the RF path might alter this (180 degrees might not be the weakest signal in all cases). As the transmitter rotates towards (640), the signal gets increasingly stronger.

FIG. 7 is a flow chart illustrating the general sensing approach to determine rotation and rotational failures. A system that is not rotating is not necessarily in a failure state. To suppress these false positives, an additional step of determining whether there is expected rotation may be required. Conversely, a system that is experiencing rotation may not need to undergo the additional step of determining if rotation is expected. Of note, the alerting of an error state would occur at 706, specifically when rotation is expected, but not occurring. This flow chart is intended to show the on-going analysis. It is not enough to just test once and be done. This is a continuous cycle as the rotator may fail at any point. This process does not need to be constant, however. The system may stop for a period to save processing power. Further, it may be configured such that the alert is only generated when the system observes a sustained failure state (706) for an extended period.

The process may be performed during a single stroke. The notion of a single stroke is a common concept in the industry, but described here for completeness. A stroke is roughly defined as the linear motion that returns to the starting point. This commonly is mapped to the bottom of stroke (through the top of stroke) and back to the bottom of stroke. It could be defined as a change in direction, top-to-top, or any arbitrary position. Critically, the use of the optional sensors, an external signal from an automation controller, or even the RF characteristics themselves can be used to identify the stroke cycle. For general rotation detection, it is sufficient to know whether the system is stroking or not. For some aspects of rotation sensing, such as torque determination, it is helpful to know where in the linear stroke the system is.

Claims

1. A method for monitoring a rod-string in a pumpjack system, the method comprising:

providing a wireless device attached to a member of the pumping system, configured to transmit RF signals to a remote receiver, wherein the wireless device moves relative to the receiver during pumpjack operation;

receiving RF signals from the wireless device at the receiver;

monitoring at least one RF signal characteristic of the received RF signals at the receiver, wherein the at least one RF signal characteristic varies as the wireless device moves relative to the receiver;

detecting a variation pattern in the at least one RF signal characteristic; and

determining an operational status of the rod-string based on the detected variation pattern.

2. The method of claim 1, wherein the at least one RF signal characteristic comprises one or more characteristics selected from the group consisting of: signal strength, signal quality, link quality, attenuation, reception strength, received signal strength indication (RSSI), signal-to-noise ratio (SNR), bit error rate (BER), packet error rate (PER), packet loss rate, dropped packet count, communication interruption rate, and retransmission rate.

3. The method of claim 1, wherein the wireless device is positioned above a rod rotator and rotates with the rod-string, and wherein determining the operational status comprises determining a rotational status of the rod-string based on periodic variations in the at least one RF signal characteristic.

4. The method of claim 3, wherein the periodic variation in the at least one RF signal characteristic results from the rod-string obstructing an RF transmission path between the wireless device and the receiver as the wireless device rotates.

5. The method of claim 4, wherein the periodic variation comprises a periodic waveform having a minimum when the wireless device is positioned such that the rod-string is between the wireless device and the receiver, and a maximum when the wireless device antenna is oriented towards the receiver.

6. The method of claim 3, further comprising:

detecting a failure of the rod rotator when periodic variations in the at least one RF signal characteristic are absent for a predetermined period, or deviate from expected historic periodic variation patterns; and

generating an alert in response to detecting the failure, wherein the alert comprises at least one of: a SCADA system alarm, a text message, an email notification, an audible alarm, a visual indicator, or an automatic shutdown command;

wherein the alert is generated based on absence of expected periodic RF variation rather than based on a dedicated rotation sensor reporting failure, and the alert distinguishes rotator failure from complete pumpjack shutdown by detecting continued linear stroking motion in the absence of rotational RF variation patterns.

7. The method of claim 3, wherein determining the rotational status comprises determining a rotational rate based on a frequency of the periodic variations.

8. The method of claim 1, wherein the wireless device moves linearly during a pumpjack stroke, and wherein determining the operational status further comprises determining whether the pumpjack is actively stroking based on at least one of:

(a) variations in the at least one RF signal characteristic; or

(b) load data from a load cell; or

(c) sensor data from at least one of an accelerometer, a gyroscope, a magnetometer, or a barometric pressure sensor associated with the wireless device.

9. The method of claim 1, wherein the wireless device both rotates and moves linearly during pumpjack operation, and wherein determining the operational status comprises:

detecting periodic variations in the at least one RF signal characteristic corresponding to rotation of the wireless device; and

detecting stroke cycle motion based on at least one of:

(a) variations in the at least one RF signal characteristic correlated with the stroke cycle;

(b) load data from a load cell associated with the wireless device;

(c) acceleration data from an accelerometer associated with the wireless device;

(d) angular rate data from a gyroscope associated with the wireless device;

(e) magnetic field data from a magnetometer associated with the wireless device; or

(f) pressure data from a barometric pressure sensor associated with the wireless device.

10. The method of claim 1, wherein monitoring the at least one RF signal characteristic comprises monitoring periodic communication interruptions or periodic packet losses between the wireless device and the receiver, and wherein the presence of periodic communication interruptions indicates that the wireless device is rotating.

11. A system for monitoring a rod-string in a pumpjack, the system comprising:

a wireless device configured to wirelessly transmit RF signals, wherein the wireless device is positioned to move relative to a remote receiver during pumpjack operation;

the receiver configured to receive the RF signals from the wireless device and monitor at least one RF signal characteristic of the received RF signals; and

a controller configured to analyze the at least one RF signal characteristic for variation patterns and determine an operational status of the rod-string based on the variation patterns.

12. The system of claim 11, wherein the wireless device comprises a wireless load cell configured to measure axial load on the rod-string and wirelessly transmit RF signals containing load measurement data.

13. The system of claim 11, wherein the pumpjack includes a rod rotator, wherein the wireless device is positioned above the rod rotator such that the wireless device rotates with the rod-string, and wherein the controller is configured to determine a rotational status of the rod-string based on periodic variations in the at least one RF signal characteristic.

14. The system of claim 13, wherein the controller is further configured to detect a failure of the rod rotator when periodic variations in the at least one RF signal characteristic are absent for a predetermined period and generate an alert in response to detecting the failure.

15. A method for detecting rotation of a rod-string in a pumpjack system, the method comprising:

positioning a wireless device above a rod rotator such that the wireless device rotates with the rod-string;

transmitting RF signals from the wireless device to a remote receiver;

monitoring received signal strength indication (RSSI) of the RF signals at the receiver;

detecting periodic variation in the RSSI as the rod-string periodically obstructs an RF transmission path during rotation; and

determining a rotational status of the rod-string based on the periodic variation in RSSI.

16. The method of claim 15, wherein the wireless device is a pre-existing wireless load cell installed for load measurement purposes;

wherein no dedicated rotation sensor hardware is required beyond the wireless load cell; and

wherein the wireless load cell transmits without any modification, addition, or awareness that rotation detection is being performed from its transmissions.

17. The method of claim 15, wherein extrema in the RSSI provide reference ticks for determining rotational position without requiring a dedicated rotation sensor.

18. The method of claim 15, wherein the periodic variation in the RSSI results from both:

(a) the rod-string obstructing the RF transmission path, and

(b) changing antenna orientation as the wireless device rotates, wherein

the wireless device antenna has a directional radiation pattern.

19. The method of claim 15, further comprising detecting a failure of the rod rotator when periodic variation in the RSSI is absent for a predetermined number of pumpjack strokes.

20. The method of claim 15, further comprising:

monitoring data from at least one additional sensor selected from the group consisting of an accelerometer, a gyroscope, and a magnetometer; and

detecting torsional energy buildup or release in the rod-string based on a combination of the RSSI monitoring and the additional sensor data.