Nov-2022

Tools for FCC unit troubleshooting

Strategies, resources, and technology can help to rapidly identify potential operational problems in FCC units targeted for upgrading and optimisation of existing units.

Warren Letzsch
Warren Letzsch Consulting PC

Article Summary

The fluid catalytic cracking unit (FCC unit or ‘cracker’) is fundamental to the successful operation of a fuels refinery. It is important that the cracker runs every day without any artificial limits. A well-trained technical staff is purposed with the responsibility to make this happen. Tools are available to troubleshoot any unit problems to provide clues to the solutions required to restore order.

A new process engineer on the FCC unit is often left to his own devices to learn the needed knowledge for the job. Table 1 is a checklist for the engineer in preparation for an FCC unit assignment. All these items are available or can be obtained during normal operations. Advanced preparation for many emerging problems can shorten the time it takes to identify the cause and make the necessary changes to alleviate the situation.

A set of up-to-date FCC unit drawings is necessary to ensure the right numbers are being used in calculations. These should be the ‘as built’ drawings verified by field inspection. Cameras are very useful since they can record the unit as it stands versus the memories of people who think they know. The computer programs for the unit, which make many of the calculations, need to be updated after every modification. The source code should be available so it is clear just what numbers and methodology are being used.

Procedures and strategies
Valid data from the unit test runs providing the material balances is a must. The yields should not only weight balance, but also carbon and hydrogen balance. Pressure balances are needed as well as heat balances to have a picture of the entire unit when it is operating correctly. These provide checks against the current values and should be compared to design numbers.

It is helpful to have the balances done at different feed rates since it is better to make comparisons at constant feed conditions. These balances need to be made after every equipment modification or when there are changes in catalyst or feed type to evaluate their effect on the unit performance. Single gauge pressure surveys are used to troubleshoot standpipe problems and can locate the source of flow issues. The instrumentation design scheme needs to be reviewed to understand the numbers.

Maintenance records for the FCC are a vital source of information. These contain a chronological list of repairs made to the cracker and their frequency. Any items that were not made during the last turnaround (TAR) should be noted since a problem can occur before the next scheduled shutdown. Data that might be pertinent to questionable equipment and locations should be monitored to quickly spot and alleviate any problem.

Inspection reports can be used the same way. Any signs of unusual wear, coke formation, or other anomalies can provide clues to the underlying causes of the problem. It is essential that the causes of recurring issues be identified as design, operational and/or procedural so that changes can be made that resolve the problem.

Identifying trends
The equilibrium and fresh catalyst analyses should also be monitored. Fresh catalyst samples can be analysed by the company doing the equilibrium catalyst (ecat) analyses, which eliminates lab differences when comparing individual values of each analysis. Catalyst samples from the wet gas scrubber, ESP or flue gas line should be taken periodically as well as from the main column bottoms to determine catalyst losses.

A separate sheet for the individual samples makes it easier to spot variations and the location of changes. Twice a month should be adequate to capture any trends and give confidence in the numbers. Sampling might be increased when issues arise.

Catalyst additions and losses should be checked and recorded regularly. Automatic loading systems can generate additions to the cracker. The losses are calculated from the concentration of the catalyst multiplied by the appropriate exit stream rate. If any equilibrium catalyst is being added, the amounts and analyses of this material should be known.

The volatility of the fresh catalyst (moisture, sulphates, chlorides, and ammonia) needs to be calculated to get the actual weight of the catalyst added. This number usually ranges from 7 to 13 wt%, depending on the final calcination received at the catalyst plant. Catalyst may pick up moisture after it leaves the manufacturer before it is injected into the unit.

Any additives should be treated the same way. The amounts and analyses of the material can help explain changes in platinum, CO index, magnesium, calcium, and phosphorus contents and are essential in calculating their retention and where they are leaving the FCC unit. All purchased ecats should be checked for compatibility with the desired operation.

Key performance indicators
The FCC unit has many variables that need to be monitored on a regular basis. Many refiners call some of these variables key performance indicators (KPIs). Table 2 lists some of these values. Each refinery has its own values, depending on what is important to its operation and needs. It is important to be familiar with the indicators and their impact on FCC unit performance to see if this may be the source of a particular problem.

Other tools for troubleshooting can be called on to provide important clues to problems and their causes. Some of these are listed in Table 3. Tracking the feedstock analyses and product properties is important since the largest single variable for any process is the quality of the material processed. Changes in feed properties will affect the yields and product properties of the FCC unit.

It is imperative that the process engineer understands the significance of all the feed tests being run so countermeasures can be taken to fix any issues with the operation. The unit test runs can be used to generate graphs showing the effect of each feed variable. The operating manual should also be checked for the licensor’s correlations.

Reaction mix sampling
Reaction mix sampling allows for direct sampling of the reactor effluent in the overhead vapour line and determining the yields without having to wait for the gas plant to come into equilibrium under the new conditions. These give the as-produced yields from the unit’s reactor. Samples can be taken every two hours when making changes to operating parameters to evaluate the impact of variations in feed rate, dispersion steam, feed temperature, and stripping steam.

Samples from the stripper can give insight into the effectiveness of stripper performance and help identify the hydrocarbons going to the regenerator. Samples taken from the end of the riser and the overhead vapour line will show the cracking that takes place in the dilute phase, cyclone system, plenum chamber, and vapour line. These tests will provide the basis for justifying a revamp to the reaction separation system.

RMS testing gives information on the feed injection system. Turndown capabilities, plugged nozzles, changing the number of nozzles, and/or the type of nozzle can all be evaluated. If there is more than one feed location, the splitting of the feed can be tested.

Lift gas injected at the bottom of the feed riser is used in some FCC operations. The amount used can be evaluated easily with RMS testing. More steam in this service, which is not effective, is reducing the overhead condensing capabilities and adding to the load on the sour water stripper. The stream used as lift gas can also be examined.

Recycle to cat crackers can be very important. Streams that might be recycled are gasoline, heavy cycle oil, decant oil, and even LCO. If separate nozzles are available, the location in the riser can be studied along with the rates of each stream. The impact of changes in the cut-points of various streams can also be explored. New units should be designed with taps in the appropriate locations to facilitate RMS testing. The testing itself is shown in Figure 1.