Maintenance gets Cracking

No one wishes for unscheduled shutdowns, but when a 13-tonne catalyst pump inside the Residual Hydrocracker Train 1 (RHC-1) at the Scotford Upgrader — part of Alberta’s legendary Athabasca Oil Sands project — failed last October, maintenance personnel jumped on the opportunity to ‘make lemonade from a lemon’.

“Since we had to change the pump, we had the opportunity to do as much turnaround work as possible,” says Sam Spanglet, vice-president of operations for Shell Oil Sands.

The Scotford Upgrader receives bitumen (high viscosity ‘extra heavy’ crude oil) from the Muskeg River mine to the north via a 493-km-long pipeline and converts it to lighter oils.

Occupying a 20-acre site near Fort Saskatchewan, just northwest of Edmonton, Scotford went into production in April 2003 and at full production produces more than 155,000 barrels per day of synthetic crude oil.

The pump failure necessitated shutting down RHC-1, with an attendant loss of production of about 35%. As serious as this was, Shell Canada, one of three partners involved in the Athabasca Oil Sands project (Western Oil Sands and Chevron are the others), was ready to turn the situation to its advantage: it immediately launched a maintenance turnaround that was originally scheduled for the spring of 2005.

Preparedness was the key, explains Spanglet.

“When we prepared ourselves in 1999 for startup in 2003, we had already established a turnaround department, even though we did not yet have a plant. The department put all the planning processes in place and had already started to plan the first turnaround for 2005; for example, understanding the equipment and the planning.

“We were able to mobilize ourselves very well, safely and efficiently. This was the first time we ever touched this unit, but we knew it inside out.”

Had there been no pump failure, there would have been 100% production last October/November and zero per cent production this spring during the scheduled shutdown of RHC-1 and RHC-2. Instead, production dropped to just 65% last fall and will again be reduced to 65% in the fall of 2005 during the re-scheduled, planned maintenance shutdown of RHC-2.

The result of being able to turn this breakdown into an opportunity was that the net loss of production due to the pump failure, although off schedule timing-wise, was zero.

Shell abides by four concepts in its maintenance planning and execution, Spanglet explains.

“The first concept we established is turnaround planning way ahead of the turnaround, and having a separate department with great planners who are totally dedicated to the turnaround. Getting involved in a turnaround is a prestigious job.

“Second, all turnaround work lists have to be submitted six to 12 months before turnaround. You cannot submit turnaround plans after this. Everything is planned: step-by-step, hour-by-hour for each job, including the number of contractors. All of the computer systems and planners are in place.

“Third, on an hour-by-hour and day-by-day basis, we measure where we are at for each job. If we start to fall behind, we have to come up with a plan on how to catch up.

“Fourth, we get the contractors involved in the turnaround a year in advance. They are engaged with us way in advance.”

Despite the fact that Shell had to round up upwards of 500 people outside its own 80-person maintenance team on nearly no notice last fall, the people came. “We had a contractor already involved who knew how to do this maintenance. It was easy to pull in the contractor on short notice,” says Spanglet.

The turnaround was completed in 39 days, six less than on the planning schedule.

As for the effect on ongoing maintenance, Spanglet explains, “By diverting maintenance personnel from other jobs, a backlog of other maintenance was created that we could live with and catch up on later. Good facilities in this business have a backlog of six to eight weeks, and we are at that.”

Shell squeezed more lemonade out of the RHC-1 shutdown at the Muskeg River mine, which had to cut back production to keep pace with the reduced Scotford Upgrader throughput.

“We also had an opportunity to enter the settlers and do preventive maintenance,” says Spanglet, “since there were many things that required so much repair because of heavy use. We did this six months ahead of time.”

Montreal-based Carroll McCormick is a senior contributing editor for Machinery & Equipment MRO. This article first appeared in the June 2005 issue of MRO Magazine.