Oct-2008

Staged FCCU main fractionator revamp

A refinery column is debottlenecked in steps to address operational concerns and strategic goals. Understanding the site specific operational nuances and developing a representative revamp plan to accommodate the maintenance plan improves the project’s success

Darius Remesat, Koch-Glitsch Canada
Michael Popowich, CCRL

Article Summary

The main fractionator of the fluidised catalytic cracking unit (FCCU) continues to be a key part of one of the most profitable units in most refineries and, typically, is the source of reliability challenges and one of the unit’s major bottlenecks.¹ Any opportunity to increase economic reliability and capacity is always strongly considered, but needs to be accomplished within maintenance plans. Fortunately, changes can be
made in steps to address various scheduling constraints and operational challenges.

Phased revamp approach
CCRL owns and operates a FCCU within the refining portion of its Regina, Saskatchewan complex. Figure 1 shows a simplified representation of the FCCU. Due to its geographic location and surrounding economic situation, there is an increased demand for refined transportation fuel products. Debottlenecking the FCCU, including the main fractionator, would help provide the supply to meet increased local demand. The main fractionator separation yields several products, including light cycle oil (LCO), slurry and a gasoline-quality product, as well as light gases for processing in the gas concentration unit. The debottleneck project was partitioned into three distinct stages:
- Revamp internals specifically to improve reliability (completed)
- Revamp internals to auxiliary equipment limits to increase capacity (completed). For example, increase capacity to the duty limit of existing pumparound exchangers
- Revamp internals and auxiliary equipment to shell capacity (for future consideration). For example, revamp exchangers, pumps, control valves and remaining internals.

Although stage three of the debottleneck is not approved, it represents the maximum potential capacity of the fractionators. This stage served as a basis from which to define the scope of the previous revamp stages to ensure they would not become a bottleneck in subsequent revamp stages. The scope of each revamp stage was also defined so as to meet the set maintenance schedules.

The original configuration of the internals in the FCCU’s main fractionator are illustrated on the left-hand side of Figure 2. From top to bottom, the tower consisted of 
11 two-pass trays, with floating valves 
in the top fractionation and LCO pumparound zone, an LCO product partial-draw collector tray, seven three-pass trays with floating valves in the heavy cycle oil (HCO) pumparound section, a HCO partial-draw collector tray, two two-pass trays with caged valves in the wash zone, seven disc and doughnut trays in the slurry pumparound zone, and a vapour feed impingement baffle plate.

First-stage revamp
The goal of the first-stage revamp was to increase the reliability of the column so consistent, on-spec product flows could be counted on for sale. Based on past tower inspections and operational challenges, the wash section prematurely fouled with coke build-up and catalyst deposits from the FCC reactor. Increases in tower pressure drop, tower flooding, temperature profile excursions and quality issues were all challenges posed by the fouling in the wash section. After a review of the entire column and auxiliary equipment, the wash section and HCO collector tray were targeted for revamp to improve reliability.

The HCO collector tray is a partial product draw with a 27 inch overflow weir. With this design, the wash zone liquid flow was very sensitive to changes in feed, and continuous flow to the wash zone was not assured. A simple modification of a given weep hole configuration in the overflow weir located above the collector tray 
deck was proposed to provide a minimum continuous flow rate to the wash zone.

The two trays in the wash zone were converted from caged valves to severe service proprietary Provalve trays. The Provalve is a fixed valve with a large round opening on the deck with a larger than the hole open area trapezoidal cap. This style of valve provides vapour push to help sweep the tray of any foulant. Figure 3 shows the Provalve trays installed in the FCCU’s main fractionator.

There was a short shutdown planned in the spring of 2007 and the HCO collector tray and wash zone tray modifications could fit within the shutdown schedule. After start-up, the wash zone operated consistently, and was able to withstand a reactor upset, which carried over  approximately 120 000 lb of catalyst fines into the fractionator. During a brief shutdown to clean up the slurry circuit from the catalyst fine excursion, the revamped wash section trays continued to perform as designed and did not suffer the effects of the catalyst fines.

Second-stage revamp
With confidence from the first revamp stage, the goal of the second stage was to increase the capacity of the FCCU to the limit of the pumparound and overhead exchangers while maintain-ing (at a minimum) the product specifications for the LCO and gasoline product. A maximum HCO draw temperature of 590°F and fractionator bottoms temperature of 680°F were maintained. The increases in unit charge, with the fixed exchanger duties, were governed by these temperature limits. The revamp simulation specifications were set based on the product specifications in Table 1.

A detailed, representative simulation was developed to allow a match of existing conditions and to provide confidence in extrapolating to the anticipated revamp conditions. The vapour feed to the main fractionator (not analysed for composition) was calculated in the simulation by product back blending. The appropriate tray efficiencies were input into the 
simulation for plant data matching.

Data from a gamma scan of the column was used to validate the hydraulic correlations of the internals to determine the actual capacity of the internals and what gains the revamped internals would provide.

Two conditions were reviewed for pre-revamp base lining and post-revamp capacity increase goals: the existing expected operation and an occasional high LCO yield case. Table 2 shows the results from the equipment evaluation for the two simulation cases based on plant data pre-revamp, a simulated high LCO yield case and the evaluation from the gamma scan. The data from the hydraulic calculations appear to support the trend information gathered from the gamma scan. Trays 9, 10, 17 and 18 experience a larger liquid and vapour load than the other trays, primarily a function of the separation profile in the column.

Based on the current equipment configuration, the main bottleneck to the increased capacity of the column for stage two appeared to be the shed decks in the slurry P/A section. The jet flood on these trays was calculated at 108% for the plant test run data and 105% for the simulated high LCO yield case. More importantly, the c-factor in this section was calculated at just at or over 0.7 ft/s for the two cases. From past experience, performance of these trays begins to deteriorate after a 
c-factor of 0.35 ft/s, with anything over 0.50 ft/s creating a large reduction in de-superheating (heat transfer). The high c-factor in the shed deck section will cause the liquid travelling down the tower to hold up in and above the shed deck section due to vapour entrainment of the liquid. Since the bottoms level is controlled by the LCO product flow, the bottoms level will decrease, causing the LCO product to be diverted back into the column. This will continue until the downward liquid overwhelms the high-velocity vapour in the shed deck section and dump into the bottoms level, causing the LCO product flow to resume. The bottoms level cycle will continue until the charge rate is reduced.