Mar-2003

Specifications — importance of getting them right

Faulty equipment specifications sometimes carry unforeseen penalties, with loss of performance. It is the process design engineer’s responsibility to meet the objectives

Steve White and Scott Fulton
Process Consulting Services

Article Summary

When process equipment does not meet the refiner’s expectations, millions of dollars of profits can be lost due to lower feed rate, less than optimum product recovery, or unscheduled shutdowns. Original equipment manufacturers (OEM) are often targeted when this occurs.

Yet all process equipment has limitations that are based on fundamental design and operating principles, which should not be violated if the equipment is to work reliably.

In some cases, the equipment selected does not suit the purpose and in others the equipment specifications result in unintended consequences that were not apparent to the designer.

It is the process design engineer’s responsibility, not the OEM’s, to ensure the equipment selection and specifications meet the processing objectives. Furthermore, because equipment is often installed in severe services such as the FCC main column bottoms system where the fluid temperatures are as high as 700ºF, available net positive suction head (NPSH) is often low, flow variability from start-of-run (SOR) to end-of-run (EOR) can be large, and the fluid contains catalyst fines and chunks of coke. Thus, even the best-designed equipment can be severely stressed.
Where possible, cost effective process flow scheme changes should be adopted to allow the equipment to operate within inherent limits, rather than simply placing blame on the OEM for poor process unit performance.

Three case studies review some common examples where equipment specification and selection reduced unit profits.

Case study 1
Low NPSH pumps
An FCC main column bottoms (MCB) pump was recently specified with an NPSH available of 8ft. Thus, a low NPSH-required pump was selected and installed. When equipment reliability was poor, the contractor and operator blamed the OEM. Yet the root cause was the design engineer’s ultra-conservative specification of the NPSH available.

If the NPSH available is low, the pump selected will always result in reduced turndown. Thus, when the refiner operated the pump over the range of flow rates required to operate the unit, the flow rates occasionally fell below the minimum. This below minimum flow rate damaged the pump due to low-flow induced cavitation.

Initially, the design engineer needs to appreciate the penalty for being ultra-conservative when calculating NPSH available. It forces the refiner to accept a pump with little operating flexibility. The selected pump will have a large impeller eye opening to reduce fluid pressure loss. Yet, this decreases the stable operating range for the pump.

As pump impeller eye diameter increases with decreasing NPSH available, the pumped fluid has a tendency to re-circulate at the entrance to the impeller. Fluid circulation in the impeller eye causes vortices that result in fluid cavitation. Operating the pump below minimum flow leads to seal and bearing failures, casing erosion, impeller erosion, and other unwanted problems such as extreme suction line vibration.

While pump NPSH required is a familiar term to most refinery process engineers, the affect of NPSH required on the impeller design and pump performance is not commonly known. Today, most pump suppliers report suction specific speed (Nss) somewhere on the performance curves. As pump NPSH required decreases, the Nss increases, and the pump stable operating range decreases.

A pump having an Nss of 18000 may turndown to only 85% of best efficiency point (BEP) before it begins to cavitate due to re-circulation in the pump impeller eye. In high-energy high head pumps, pump damage can occur rapidly and maintenance cost can be very high.

Although selecting high Nss pumps and the resultant narrow operating range is known to rotating equipment engineers, process designers need to understand the penalty associated with specifying an ultra-conservative NPSH available. In some instances, low NPSH pumps must be selected, but in those cases where it is absolutely necessary, design engineers need to ensure the process flow scheme will give operating personnel the flexibility to maintain the pump flow within a narrow range without adversely affecting the FCC unit operation.

MCB system review
New pumps are frequently installed during revamps where they must fit into an existing process system and operate properly within constraints. Therefore, the designer does not have a “clean sheet” of paper and must find cost effective solutions working with the existing equipment. When trying to install new pumps, plot space will determine location rather than ideals, such as minimum suction piping run.

Figure 1 (previous page) shows a typical FCC main column bottoms (MCB) system. Reactor effluent enters the column at temperatures of 980–1015ºF where the MCB system must remove up to 35% of the heat so the reactor products can be fractionated. Fluid mixed with catalyst and coke fines is withdrawn from the bottom of the main column and pumped through heat exchangers, then back to the column as sub-cooled pumparound return (PAR) and quench.

PAR liquid flows down the column through internals such as shed trays or grid where heat is transferred from reactor effluent to the PAR liquid. To prevent coke formation, most refiners maintain a constant temperature in the bottom of the main column by varying quench flow rate. However, MCB circulation rate depends on the system design, and the operating philosophy can cause large flow variability from start-of-run (SOR) to end-of-run (EOR).