Apr-1998

Reducing the danger of maintenance exposure

Maintenance cost overruns result from inadequate cost estimating and the emergence of unidentified work that has to be done

Andrew W Sloley, Process Consulting Services

Article Summary

Plant turnarounds often end with the maintenance work over budget. Such budget excesses can arise from additional work that had not been foreseen, from extended shutdown length and from expediting charges. Unplanned work increases costs and reduces profits; and in a climate of relentless cost cutting, reducing unplanned work  is an important step in keeping plant costs under control.

Business (cost) exposure in maintenance work arises from two areas. First, the work may have been planned at lower than the actual cost. Second, “surprise” items are found after the equipment is opened. Cost control can work in either case. In the first area, costs must be properly estimated. In the second, maintenance items should be identified in advance.

Most advice in the literature on improving maintenance costs and maintenance control emphasise reducing the cost of the identified work and on estimating its cost correctly. While useful, this step is not as important as knowing in advance what you need to fix. We will concentrate on identifying problems in advance.

Once a work item has been identified, the exposure in the maintenance budget to cost overruns is a function of how well the repair has been defined, procurement for goods and services, market conditions and estimating procedures. With good control in these areas, costs can usually be maintained within 20 per cent of projected values.

In comparison, missing an item completely has a much larger affect. First, you miss the cost by 100 per cent on the base price. Second, procurement must be expedited – at costs that are often 1000 per cent or more of standard procurement rates. Third, and worst, either by the nature of the work itself or its impact on other activities, the entire turnaround can be delayed.

Defining maintenance requirements is the responsibility of the entire operations staff. This includes support engineers as well as the unit operators. Simple methods are available to identify problems in advance with most process equipment. For our examples, we will concentrate on towers (distillation, absorption, extraction), heat exchangers and drums. Similar simple methods are available for other process equipment.

Getting experienced process engineers involved in equipment and unit surveys is one key element. One type of survey is to take pressure and temperature profiles across the process equipment. Another type is whole unit test runs. Test runs immediately after startup set a base line of unit performance. Follow-up test runs during the unit operation gather data for comparisons.

Comparing test run results over time is a key to identifying equipment performance problems. Identifying the cause of equipment performance failures is the key to proper definition of maintenance requirements.

The following examples show different types of field data that can identify some common equipment problems. The examples concentrate on how engineers, using basic, straightforward principles and applied thought, can identify major problems. Thousands of pages of examples could be used to illustrate different approaches. The examples are only a taste of what can be done. The engineer’s field experience, ability to grasp the basics, and imagination in applying them are the only limit.

Tower maintenance
Tower maintenance includes both shell (corrosion) repairs and internals repair. Shell damage can often be detected in advance by appropriate thickness testing. Internals damage is more difficult to detect. The major unexpected problems in tower internals are 1) damaged equipment and 2) corroded equipment. High tech approaches to evaluating tower internals for suspected damage are sexy but often unnecessary. Simple methods with temperature probes and pressure gauges can identify many problems.

The advantages of temperature and pressure measurements are:
Cost – they are cheap compared to high tech methods.
Speed – results are obtained immediately.
Simplicity – special arrangements for radioactive materials or methods are not required.
Accuracy – results from gamma scanning and radioactive tracer methods can be very difficult to interpret. By comparison, results from temperature and pressure measurements are easy and straightforward to interpret.

Temperature and pressure profiles give useful results for two reasons. First, temperature, pressure and composition define a process system. If you know two of them, you know the third. Second, pressure drop is required to drive fluid flow. Fluid flows from high pressure to low pressure.

Example one
Stripping section analysis
Figure 1 shows a vacuum crude tower. The crude tower prepares feedstocks for downstream lubricant base-stock processing and bitumen production. Light material in the residue has been causing off specification penetration on the bitumen. The refiner is attempting to move to a six-year run length on this unit. If necessary repairs are not made later in the year, six years may elapse before a second fix attempt is made.

This unit has a history of stripping section damage in the bottom of the tower. A properly functioning stripping section significantly reduces light (vacuum gas oil range) material in vacuum tower bottoms. This affects both HVGO yield and residue quality. In spite of the history of damage on the unit, refinery management wants assurances that any money spent on advance procurement of replacement trays is really required. What assurances can be delivered?