Improving solids handling in a delayed coker
A reliability model applied to a coke handling system demonstrates the benefits of compliance with a preventive maintenance programme
Randy Hull, Steve Stewart and James Teets
Viewed : 4400
Petroleum refineries are increasingly adding cokers and hydrocrackers to meet refining challenges from heavier/more sour feedstock and increasing demand for higher distillate/
gasoline ratios with very low sulphur levels. Uplift margins on US Gulf Coast cokers have moved to over $30 per barrel, so that cokers are among the most important units in a modern complex refinery. Unfortunately, these new process configurations create very high volumes of solids. A typical large-scale delayed coker produces well over 1 million t/y of solids, which would fill 30–35 railcars in a day. That is a lot of solids handling, which, for most refiners, is not a core operational competency.
Interruptions in coker units are widespread due to unplanned outages in the industry as a result of breakdowns in the solids handling systems. Shortening coke drum reheat cycles to compensate for downtime can lead to premature drum cracking, resulting in significant downtime and lost profitability.
Marsulex has developed a total maintenance and reliability model for delayed coking units (DCU) to help refiners plan a preventive maintenance strategy to significantly reduce the total cost of ownership for coke handling. This model maximises a coker’s profitability by reducing equipment repair and replacement costs, as well as minimising unplanned outages. Refiners who choose to carry out their own solids handling are encouraged to adopt a similar total maintenance cost approach.
Since most refinery operations are characterised as continuous processes handling liquid and gas streams, handling solids in a batch process operation is usually not an area of excellence within refinery operations. Effective performance of the coke handling and batching operation requires effective performance in safety, operations and reliability.
Operations know-how and experience constitute the second critical area for the coke cutting business. The coke drum heading and unheading systems, the coke pit and the coke conveyance areas are highly automated with solids handling technology that requires well-trained operators.
The third area is reliability, which is the focus of this article. Since the coke cutting and movement business is so mechanically intensive, this is an area of high cost with many opportunities for cost reduction. Reliability separates the competition in DCU solids handling and maintenance; however, with the constant pressure to reduce costs, eliminating reliability programmes is a short-term fix with long-term consequences. It is Marsulex’s philosophy and experience that reducing the reliability investment — namely, scaling back preventive maintenance and systems investment — can be an expensive decision in the long run. The “acquisition cost” dollars saved when bidding on a services contract are insignificant when compared to the total lifetime cost from unreliable, poorly maintained equipment. A world-class preventive maintenance (PM) programme is key to delivering a successful operation, whether a refiner chooses to self-perform or elects to outsource these services.
Marsulex’s philosophy of reliability is best described by the Maintenance Parabola shown in Figure 1.
The Maintenance Parabola is an established but subtle concept. The curve shows the relationship between PM expenditure and total maintenance cost. Starting at the left of the curve and moving right, as PM activity is increased maintenance costs are reduced, significantly at first. As PM expenditure increases, moving right on the curve, maintenance cost reductions reach a point of diminishing returns. To the right of the curve, minimum total maintenance costs begin to rise with increasing PM expenditures. Marsulex’s philosophy is to stay focused on delivering an effective PM programme to model the above relationship. The programme optimises PM activities to deliver the lowest total maintenance cost, which includes total preventive and corrective maintenance, as well as consequential opportunity costs related to lost production experienced during unplanned outages.
Equipment failure background
Marsulex categorises repairs, or failures, into four basic types:
• Type I: planned or expected failure
• Type II: unplanned premature failure
• Type III: incident
• Type IV: wreck.
Type I, a planned or expected failure, is rarely complicated and can be assigned a repair cost of 1X. Type II, an unplanned premature failure, has a cost that typically averages 3X, or three times that of a Type I failure. A Type II failure occasionally has additional costs, such as an operational outage. A Type III failure, considered an incident, is a premature failure with a typical cost of 6X. Type III failures often involve an operations loss, or a logistics cost to avoid an operations loss. Type IV, a wreck, is the most severe failure. Type IV failures average 9X or greater repair cost versus the Type I planned/expected failures and can include significant premature failures, operations or logistics costs and insurance claims.
Figure 2 shows a typical failure distribution as a function of PM compliance levels, which were estimated in the development of the reliability model.
The values vary by equipment type, but the trend is consistent: increased PM compliance yields a higher distribution of Type I failures. Conversely, decreasing PM compliance, or an ineffective PM programme, yields an increased percentage of Type II through Type IV failures.
The objective of an effective PM programme is to reduce costly and unplanned failures by increasing the less expensive preventive and planned maintenance activities.
Add your rating:
Current Rating: 3