Drilling Rigs Evolve Seeking Productivity Gains

The vision of fully automated drilling rigs driven by big data gathered in real time looks so far off but there are people working on a road map to help the oil and gas industry find its way there some day.

Drilling rigs working onshore are evolving, driven by the need to manage the high cost of mass producing thousands of wells needed for unconventional development. The rate of penetration (ROP) is the most commonly used measure of efficiency, but the location and the quality of the hole can be significant variables.

The number of wells drilled per year is up as the rig count goes down with many older rigs finally being retired. The rig market overall looks soft, but demand remains strong for late-model rigs wired to handle computer controls (AC drive). While they command higher day rates than older rigs, by drilling wells in fewer days they can help oil companies lower the cost per foot drilled.

At a panel discussion at the recent IADC/SPE Drilling Conference and Exhibition in Fort Worth, Texas, Michael Power, manager of unconventional resources for global drilling and completions at Chevron, said the company is seeking to reduce the average cost of drilling onshore wells in unconventional formations by 15% or more a year.

The goal has been regularly reached with a combination of incremental improvements. In addition to the rate of penetration, gains can be made in other ways, such as reducing nonproductive time, or the time and trucks needed to haul a rig to a new site. “A level of automation speeds drilling and movement of rigs,” Power said.

One indication of the pace of change is the growing market share of Helmerich & Payne (H&P), whose rigs now represent 15% of the onshore market in the United States—about twice its share in 2001—and the company is adding at least two rigs a month to its fleet.

Its rise dates back to its decision to be the first in the industry to design and build a new generation of drilling rig called FlexRigs. A critical difference is those rigs were wired to accommodate systems from the company automating certain functions, increasing a driller’s productivity. It has built five generations of the rigs and is working on the next.

Significant drilling time reductions have been a trend in unconventional exploration. H&P has drilled 10,000 ft of hole in 10.5 days from the start of work on a horizontal well to when the rig is released, said Jeff Flaherty, senior vice president of land rig operations at the company.

But gains in the rate of penetration are getting harder to find. “We are moving to the short end of the bell curve on the time to drill,” he said

Ever faster drilling takes its toll. “Unconventional drilling is hard on equipment and hard on people,” said Flaherty, who said the cost is going down per foot drilled, but the wear and tear is squeezing the company’s profit margins.

In a panel discussion put on by the SPE Drilling Systems Automation Technical Section (DSATS) about what it will take to sell the next generation of data-driven drilling tools, Brett Borland, manager of drilling engineering for global wells at ConocoPhillips, said those selling new technology need to look beyond ROP as a measure of value when selling things that usually cost more and come with the risk that they will not work.

“Selling based on drilling faster will not necessarily get you enough” added value, said Borland. “Going from drilling 90 ft/hr to 150 ft/hr may not be enough meat on the bone to sell it.”

ConocoPhillips is testing whether there is a significant reward for reducing “drilling dysfunction,” such as excessive vibration while drilling or sticking and slipping, both of which can destroy drill bits and other components in the bottomhole assembly, halting drilling while costly broken hardware is replaced.

ConocoPhillips is currently running a five-well test in south Texas using drillpipe wired with coaxial cable to provide a real-time stream of downhole data to guide digital controls from National Oilwell Varco (NOV), with a goal of increasing drilling efficiency. Borland said it has already offered valuable insights, which will change how the company drills a variety of wells. Next, it will be used to test the value of NOV’s computer controls plus a large data stream.

An important element of the plan for ConocoPhillips was a system to motivate workers on site to use the new control system while drilling. “All those on the drilling crew are incentivized to make it work,” Borland said.

Getting Past Rocks

The evolution of drilling technology is similar to the continuous improvement process seen as manufacturers embraced lean approaches to pare mass production. Overcoming one barrier to productivity highlights others.

“When you lower the water level, new rocks appear,” said Chris Marvel, senior vice president at Janus Consulting Partners, whose career has focused on using automation and data to increase productivity. When it comes to drilling, the rocks appearing are opportunities for productivity gains, ranging from easier operations based on clever designs to control systems allowing greater use of computer-controlled components.

A new drilling rig from Schramm, a drilling rig maker hoping to grab a piece of the shale drilling market with a heavier duty rig—the T500XD rig has a 500,000 lb hookload—is an indication of the range of options facing the industry.

On the design side, its new rig has an elevated drilling floor leaving an open space below that is supposed to reduce the time needed to install a blowout preventer, and components that can be shipped on 10 truckloads—about half as much as comparable sized rigs.

On the technology side, it is an AC rig wired to make it easier to install electronic controls, with hydraulic systems controlled by programmable logic controllers, and a central data collection and analysis system.

“We took a really hard look at where the industry is going and what we needed to build to fit future needs,” said Peter Christian, vice president of oil and gas equipment at Schramm. “The foundation of the technology is that it is really easy for other people to plug into our control system.”

It can accommodate an automated pipe handling system that moves workers off the rig floor into a climate-controlled booth, and a system that allows the rig to walk on drilling pads from well to well. Walking rigs are becoming a standard item. Christian said Schramm’s stand out because they turn by 5° for each step taken by its hydraulic feet.

Looking ahead, Christian said one of the obstacles to productivity the industry wants to get past is the need for hiring directional drilling consultants to guide drilling through the curve from vertical to horizontal. “They are trying to find a way not to have to hire a directional driller. That gets really expensive,” Christian said. Schramm exports 75% of what it produces, and in other countries those skilled hands are hard to find.

Christian said the focus needs to be on maintaining a steady pace that minimizes time lost for fixing problems. “People get hung up on rate of penetration. We can go so fast we outrun our ability to get cuttings out of the hole,” he said.

While the talk about drilling productivity is focused on cost reduction, the potential return on investment from drilling wells that produce more oil and gas can be far larger.

Halliburton and Devon Energy recently used extensive subsurface reservoir testing to create a field development plan that turned a marginal section of the Barnett Shale into a profitable liquids-rich development. The method used to create a more productive network of fractures also saved money by using considerably less water and sand.

A critical part of the plan was moving the average later section down 60 ft in the formation. The reservoir models showed that change would mean more productive wells, but would also increase the risk. If the horizontal sections were below the planned depth, there was a greater danger of fracturing into the formation below the pay zone and flooding the well, which would ruin it.

“We need to do more to understand the subsurface, to target the most prolific section of the formation, and complete it most effectively,” Karl Blanchard, vice president of production enhancement at Halliburton, said during the conference.

Data Inaction

After years of working on automation systems in and out of the oil industry, Marvel is now focused on using data better, combining structured data—numerical measures—with unstructured data, such as who was in charge of drilling at the time. There is an opportunity to improve, Marvel said, because “we are not using the data we have.”

Often, drilling data is never looked at after the well is completed, because with more wells being drilled faster, engineers supervising multiple projects do not have time to look back over past jobs.

When the question was raised during the DSATS discussion: Is the industry doing a good job using the data it gathers now? The answer was: “Absolutely, we do not. We have not scratched the surface of it,” said Shahab Mohaghegh, a petroleum engineering professor at West Virginia University.

Recently Mohaghegh, who is known for using methods such as artificial intelligence or neural networks to analyze data, has been developing a system able to predict drilling problems before they occur. So far, it has been tested using drilling data gathered from wells drilled by a large oil company.

There is already an early warning system based on drilling data that is in use on drilling rigs. Sekal is marketing software based on more than a decade of work at the International Research Institute of Stavanger in Norway. The model allows Sekal’s software to do “engineering while drilling,” comparing observed data with a constantly adjusted range of safe values to see if problems are developing.

The next step for Sekal is a “system to take over when the driller takes action that can put a well in an unsafe condition,” said Bill Chmela, vice president for the Americas at Sekal, in a presentation at the conference. He said the program, which has performed reliably in simulations with drilling data, is likely to be tested for the first time later this year.

It is a difficult sell because the industry is reluctant to accept machine control. “You need to do it in baby steps,” Chmela said. “You need to get comfortable with each step before the next one.”

Another data analysis company, which has worked for the US military, Ayasdi Government Services, is now seeking data analysis work in oil and gas. It seeks out hard-to-detect patterns in data. The amount does not need to be large, said Benjamin Mann, vice president of energy at Ayasdi. “Big data is a big cliché,” he said. As for the data speed required: “It is not real time. It is in the time we need to use it,” he said.

Automated Drilling Fluid Monitoring Seeks Lower Costs, Less Time Lost

A handful of new ways to continuously monitor drilling ­fluids are being offered as a way to improve drilling performance. Backers say more and better measurements can lower the cost of the mud used to create heavier, viscous drilling fluids by allowing more precise adjustments, and reducing the time lost due to problems caused by poor drilling fluid quality.

To turn their research and development efforts into a business, they will need to convince those involved with drilling that they are missing something. “My understanding is mud properties are tested a couple times a day,” said Steve Chackowicz, executive vice president at Aspect Imaging, which recently rolled out a new device for continuously testing drilling fluids. “They do not know what they do not know.”

There does appear to be industry interest in learning more. Aspect was one of two companies introducing new continuous fluid monitoring devices at the recent IADC/SPE Drilling Conference and Exhibition in Fort Worth, Texas. Representatives from Aspect and Ultra Analytical Group said they gathered a lot of business cards from potential users at the meeting.

At the conference, Petrobras reported on its program to develop a continuous drilling fluid monitoring system that it hopes will lead down the road to an automated mud management system. Not far from the conference site in north Texas, two drilling rigs are using a device from Ultra Analytical providing measures of drilling fluid density every 3 seconds from sensors inside a steel tube submerged in a mud tank.

In south Texas, another startup company with a new drilling fluid monitoring system, Onsite Integrated Services, was installing the equipment for two companies, ConocoPhillips and Halcon Resources. Onsite’s founder, Jason Norman, said he has been hearing from others. “There is more buzz in the last 3–5 months in my inbox than I have ever seen,” he said. “I have been pinged from all over world.”

Norman has been working on ways to improve drilling fluid management for years while working for two large oil companies. Now that he has started a company, Norman is focused on tracking how the fluid changes from when it goes into the hole to when it comes out by doing mass balance analysis. Those calculations are compared with conventional calculations based on drilling data. The analysis can indicate how efficiently it is clearing out cuttings from the hole, whether fluid is being lost in the formation, or whether there has been an influx of fluid or gas from the formation. Based on his past field testing, he would expect some observations that will challenge the common wisdom. “We are going to discover things that we never knew existed,” he said. “We will look at it (data) and say, ‘Huh, that clearly is not what we thought was happening.’ ”

These devices also represent a change in who sees data and when. These electronic monitors are constantly testing and sending data online to all those involved in drilling a well.

The Ultra Analytical device is being used to drill in the Barnett Shale in north Texas. The drilling consultant representing Vantage Energy on the rig, Pat Blackmer, said getting updates every 3 seconds of the mud weight on his iPhone, even when he is home in Colorado, has been valuable.

“There is a margin of human error,” he said. With periodic testing, the error may be in not testing at moments when the data would reveal a problem. Constant mud weight measurements have revealed “light spots” in the mud that can allow gas and fluids from the formation into the well. He said that data has allowed the crew to better manage the mud weight, reducing the number of times drilling needs to be interrupted to deal with kicks—generally small well control problems that might become big ones.

Robert Sickels, the inventor and a vice president at Ultra Analytical, said constant observations of the mud weight in the fluid allow a different approach to drilling fluids management. The constant flow of data from two locations—the tank where it collects after cuttings are removed, and from the tank holding fluids to be injected—allow regular, small adjustments in mud or fluid levels. “We think we can save about 20% in the cost” of the mud, Sickels said. Larger savings are possible if more consistent fluid quality management can reduce time lost fixing problems, many of which are connected to fluid management.

The Onsite analysis measures more things to track how effectively the fluid is clearing the drilling cuttings out of the hole—rapid drilling can lead to buildups that can slow drilling or stop it altogether—and offers a check on whether costly remedial actions are really needed, Norman said.

Better, regular hole cleaning monitoring “can reduce the days on wells with faster cleaning, and more efficiency and more consistency,” Norman said.

Tough Competition

Constant fluid monitoring is competing with a cheap and simple status quo. One of the most widely used density measures is a simple combination of a cup holding a set amount of drilling fluid and a scale, known as a mud balance.

“The idea is a simple beam balance and a known chamber,” said Chris Marvel, senior vice president at Janus Consulting Partners, who previously worked as a consultant developing the continuous test systems. “Anyone can make this device (mud balance) give a good reading. No special talent is required, and it will still work after someone uses it for a hammer.”

When Sickels heard from friends how mud weight was measured, he decided there had to be a better way. Finding one proved harder than expected. It took him 3 months before he got the idea to measure fluid density based on the pressure exerted by the fluid, and found an electronics supplier near his home outside Fort Worth with a sensor capable of doing so.

The resulting device has a handful of sensors inside a metal tube that is about 36 in. long, which sends out readings every 3 seconds over a connected line. It is shipped in a case that weighs 90 lb, and the heaviest part of that is the box, he said.

The FlowScan device from Aspect also does continuous testing, but the similarities end there. It can measure many more properties—shear stress, shear rates, and plastic viscosities are included—and it weighs about 2,000 lb.

Drilling fluid density is at the top of the list because it is essential for well control. There is also interest in the salinity for analyzing cuttings; water cut to determine the oil/water ratio; and rheology for measuring hole cleaning efficiency, calculating the impact of tripping pipe on well control, and analyzing fluid dilution.

“Essentially rheology is the study of fluids and how they flow. We are in this area,” Chackowicz said. “Whether it is sausage being pumped through pipe for hot dogs or ketchup, or any other material with fluid properties.”

Based on input from interested customers in the oil business, Aspect is working on extending what the FlowScan device can measure by magnetic resonance imaging (MRI). What makes it unique is its size—a cube with sides about 3 ft long compared with room-sized MRI machines found in hospitals that consume far more power.

Norman is interested because this could help him reach one of his goals: measure more parameters with fewer instruments. In this case, it could be up to eight parameters with one machine, but he said the cost of the FlowScan and the size are both significantly greater than the coriolis meters he is using for mass balance analysis.

Routines Reconsidered

Norman thinks more drilling fluid information can pay for itself by allowing a critical look at costly maintenance routines, such as sweeps. These fluid injections to clean out cuttings, which can cost USD 5,000 per sweep, are frequently done on a set schedule, with some rigs doing a sweep every time a stand of pipe is added, or every other time. It is hard to judge if these remedial treatments are worth the cost, because hole cleaning information is lacking, he said.

Norman said the mass balance analysis should provide more accurate readings of the makeup of the drilling fluid as it leaves the well and how efficiently it is clearing out the hole, providing an objective measure of when a sweep is needed.

“The bottom line is, we as an industry do not do a very good job managing the things we cannot clearly quantify. Hole cleaning efficiency is just one topic of hundreds that we do not have a clear cut way to conventionally quantify what is really happening downhole,” he said, adding that, “The end result is that we continue to perform remedial hole cleaning methods on a preventative maintenance schedule, because we really don’t know exactly what is happening.”

Delivering the value from constant monitoring will require new software. Onsite has partnered with RigMinder to develop a program to analyze data by combining the mass balance numbers with drilling performance measures, such as torque and drag, to highlight how conditions are changing in the hole based on a broader look at the available data.

Experience shows the barriers to change are not just financial. A 2010 technical paper reported on a field test by Halliburton using automated mud monitoring equipment. Marvel, who was an adviser on the project, said the test showed the system worked but it did not lead to any business.

One barrier is the business model of the companies selling the mud used in drilling fluids. The organizations were created to make money selling materials, so it is hard for them to justify investing in a different business model based on offering a service, Marvel said, adding, “It was a little too early and there was a little bit of resistance by the field people to adopting it.”

The comment about the crews echoed a regular refrain of those working to introduce new controls on drilling rigs; nothing changes unless an innovator can convince workers on the rig they will benefit from a change. When Sickels took his device out for a first test, the mud engineer said, “If this thing works, I am out of a job.” Sickels responded, “No, you are going to be running this thing.”

In practice, Sickels said crew members have embraced the tool. While he is happy it is being used, he worries about mud engineers failing to continue to conduct periodic tests to ensure the electronic device is delivering accurate data. It is new device that is just going into commercial service. There are conditions that can throw off its readings. For example, if water-based mud is used it needs to be regularly cleaned to remove buildup on the outside that could interfere with the pressure measurements it uses to calculate density.

Machine Control

The commonly made observation that every well is different irks Norman. He said that after so many wells have been drilled, the potential problems that can be encountered are known, and what makes wells different are the decisions made by drillers when encountering those conditions. Different drillers react differently to the same situation. Some miss the signals completely.

For Blackmer, continuous monitoring helps reduce the variations that come from crew turnover. While a mud engineer is in charge of controlling what goes into the drilling fluid each day, the execution of the plan and the periodic testing is done by a member of the crew, who does that when he is not busy controlling the derrick. That means he is unable to take measurements during activities, such as pulling the drilling fluid properties.

Some crew members are more diligent than others about following instructions. Rather than adding a bag of lime over 3 hours according to the daily plan, a crew member might dump it all in at once. Blackmer said mud density monitor “will tell you if they are mixing in products too quick or not mixing in enough.”

The value of detailed drilling fluid data will depend on how well it is analyzed and presented. “In 3 years’ time, everybody will be doing this stuff,” to measure drilling fluids, Norman said, adding that differences in performance will be based on the quality of the model used to interpret data and make decisions.

For example, one of Norman’s goals is to use flow rate trend data to predict mud pump failures—slowing rates indicate wear—so that replacements can be planned with a minimum of lost drilling time.

Petrobras’ research center in Rio de Janeiro is working along with experts at the nearby Federal University of Rio de Janeiro to create an online fluid’s testing system, which it hopes will be able to detect and react to potential problems sooner.

The first half of the 7-year project has been to find a way to reliably gather multiple streams of data on drilling fluids as it comes out of the well, and at the point it is injected into the well. The ultimate goal is a computerized system to create a more detailed measure of drilling fluid properties that can be used to automate the adjustments needed to restore the quality of the mud to the specifications needed for safe, fast drilling.

For the past 4 years, the work has focused on identifying and adapting devices able to measure multiple drilling fluid properties as it flows by. It is seeking measures of fluid density, viscosity, mud weight, solids concentration, salinity, and the stability of oil-based and water-based fluids.

The next stage will require improvement of measurement devices leading to a field test. Ultimately, it hopes to create a system that is reliable and durable enough to use the data to automatically adjust drilling fluids while drilling.

Norman sees the value of that someday, and has considered how to create an automated system as capable as a good mud engineer. For now, he said the plan is to target a few problems: “Let’s do what we know works to add value, and get everyone excited about it.”

For Further Reading

SPE 167978 Development of On-Line Sensors for Automated Measurement of Drilling Fluid Properties by Sergio Magalhaes, UFRRJ et al.

SPE 168018 Safer Tripping Through Drilling Automation by Bill Chmela, Sekal, et al.

SPE 137999 Making Real Time Fluid Decisions with Real Time Fluid Data at the Rig Site: Results of Automated Drilling Fluid Measurement Field Trials by Shawn Broussard, Halliburton; Chris Marvel, Long Rider Solutions, et al.

SPE 168007 Real-Time Operations Support for Geographically Dispersed Operations by Richard Kucs, OMV; John Thorogood, Drilling Global Consulting Consultants, et al.