Drilling advances

Installation of the first top drive system, on a jackup in Abu Dhabi some 37 years ago, was intended to signal the death knell for stand-limited kelly drilling, shelving it alongside rotary phones and paper road maps. Since that 1982 debut, electric and hydraulic top drives are as ubiquitous on a drilling rig as head office complaints over cost overruns.

Generational upgrades notwithstanding, precipitous top drive failures occur all-too-frequently, and when they do, any ambition of sticking to the program budget typically goes out the window. In his 2011 Masters thesis at the University of Stavanger, Andrian Rognerud, now technical manager for Norwegian marine insurer Skuld Offshore, put the top drive among the leading culprits of hardware-related downtime. Consequently, it should come as no surprise that results of a top drive survey conducted more recently by researchers at the University of Texas at Austin (UTA) show that reducing downtime and improving the ability to predict failures before the system goes “kaput” top the industry’s wish list, far surpassing the desire to increase torque or tonnage capacity. The operator and contractor survey landed the top drive hydraulic system at the top for relative likelihood of failure, followed not-so-closely by the electrical power system, rotary seal and control system.

Obviously, a top drive failure, at best, puts the brakes on drilling new hole, leaving the crew standing by, awaiting either a repair or replacement, and watching as costs mount. Over 40% of the survey respondents said a top drive-induced NPT event lasts between 12 and 24 hr, with a single event on a land rig costing between $60,000 to $120,000/day, exclusive of labor, parts or a replacement system.

In an industry with redundancies out the yazoo, the typical tally of wellsite consumables does not include a spare top drive, lying in wait aboard the supply vessel or by the pipe rack, thereby accentuating the need to ensure the system remains in a fully functional condition. That assurance used to be wholly at the mercy of periodic inspections by preventative maintenance technicians, but given today’s digitally-infused environment, machines can now tell when other machines are ailing.

AI predictive maintenance. For instance, SparkCognition of Austin, Texas, which has built artificial intelligence systems for industries as diverse as aviation and financial services, sees AI-powered predictive maintenance as the hands-off antidote to untimely failures of top drives and equally critical components. As the company explains, predictive maintenance capitalizes on machine learning to predict top drive or other component failures, well before the rig has to be shut down. “An unsupervised learning model collects data from sensors on the rig, monitors and analyzes the sensor data, and raises an alert for impending failures far in advance, thus vastly decreasing asset downtime,” according to SparkCognition.

The continuous learning and self-refinement intrinsic to an AI-centric preventative maintenance system analyzes patterns to identify failures, even those never before encountered. “This flexibility also allows an AI solution to catch unusual or so-called ‘edge’ cases that a rules-based approach cannot. It detects impending failures, long before any traditional approach can,” the company says.

Condition-based monitoring. Meanwhile, across town, scientists at UTA’s Rig Automation and Performance Improvements in Drilling (RAPID) group have been involved heavily in work aimed at enabling condition-based monitoring (CBM) of top drives, largely using data that rig-based sensors were already collecting. After considering three distinct top drive health monitoring techniques (thermal, vibration and oil analysis), the university researchers settled initially on developing a thermal model in tandem with a fault detection algorithm. The model not only proved very effective in monitoring the lubrication and electric motor subsystem of an AC-powered top drive, the thermal analysis method can be implemented with the sensor suite already available on most contemporary top drives, says RAPID Senior Research Scientist Pradeep Ashok.

“Our first aim was to put additional sensors on the top drive, but then we realized there already was a lot of thermal sensors that were collecting data all the time,” he told the IADC Drilling Engineering Committee (DEC) Technology Forum last November in Houston. “So, we shifted our initial approach from putting more sensors on the top drive to using these data to see, for example, if the motor or gear box lubrication was degrading. We build a thermal model for these, which actually predicts what the temperature needs to be, which is compared to actual data.”

Thermal monitoring leverages the motor winding and gearbox oil temperature sensors already installed on most top drives. An estimation of model parameters from real-time temperature measurements, in turn, can determine whether the estimated parameters are indeed healthy. “With thermal modeling, we can prevent these failures from happening,” Ashok said.

At the time that RAPID researchers began their response to increasing incidents of top drive failures, the industry was keeping capital in a tight fist. “When the project came out in 2015, oil prices were low, and nobody wanted to spend any money. Basically, we had to see what we could do about CBM without spending any money and how best could we use the sensors that are already present,” he said.

At last count, no vibration or fluid monitoring sensors have yet been installed, while the thermal model has been put in place on two rigs. Ashok said developing and installing an entirely new suite of sensors would cost around $15,000, far less than the costs of even one NPT event. “You would still save a whole lot in the long run,” he said. 

About the Authors

Jim Redden

Contributing Editor

Jim Redden is a Houston-based consultant and a journalism graduate of Marshall University, has more than 40 years of experience as a writer, editor and corporate communicator, primarily on the upstream oil and gas industry.