In the past two decades, technical manpower and average experience levels within the oil industry and its affiliated engineering, contractors and technology companies declined as the need for lower operating costs drove most decisions. Senior experienced staff members took early retirement or were displaced with younger professionals. Supported by new state-of-the-art technological tools, but lacking adequate training and mentoring, these younger workers are expected to deliver more and faster. At times, designers and engineers are working under rush deadlines and lean supervision, which can lead to frequent accidental engineering missteps, faults and miscalculations. This article illuminates the current industry situation and the need for proper training and mentoring, and a broader and deeper experience base. Several examples illustrate poor design and engineering encountered during a recent refinery project.
The following examples illuminate the present condition with the energy industry. These incidents were encountered during outside engineering work on just a single project at Alons Big Spring Refinery (BSR). Fortunately for BSR, the refinerys core project team of three membersa project/mechanical engineer, a refinery process consultant (RPC) and a process control/electrical engineerall supervising this project are experienced and versatile enough to quickly catch and correct all the problems. The RPC also interjected new thinking, and the refinerys team was able to achieve a very successful project outcome.
Fig. 1 shows an overview of the recent hydrotreater project at BSR. The unit feedrate was being increased by more than 25%, while product specifications were stringently tighter. The new equipment included a liquid-phase upflow reactor No. 1, a splitter, stripper product pumps and associated new heat exchange train and piping. Additionally, the existing vapor-phase downflow reactor, compressor, hydrogen heater, stripper reboiler, and airfin coolers and piping were being revamped. The hydrogen heater provides primary preheat for the feed to reactor No. 2. The reactors, the heart of a hydrotreater operation, can directly impact plant profitability through any compromised design or reliability. A semi-regenerative reformer is the refinerys sole hydrogen source.
| Fig. 1. Project overview for new hydrotreater at BSR. |
New thinking for design.
In this project, new thinking changed old, established reactor design practices for increased operational flexibility and economic advantage. A common industry practice is to design the units reactor and heat transfer equipment, including the heater(s), based on:
a) Both reactors being at the start-of-run (SOR) and/or both reactors being at the end-of-run (EOR), in tandem, based on a four-year run length and
b) Average hydrogen purity, a value of 80.2% for BSR.
But, for enhanced operational flexibility, the RPC asked that other scenarios be taken into the design of the equipment to cover a) staggered reactor operation, with reactor No. 1 being prematurely at SOR while reactor No. 2 continues to run its course and vice versa, and b) the expected 74%88.6% range for hydrogen purity.
The revised basis increased the sizes of a reactor and heat exchange equipment, as well as the sizes of the hydrogen heater and reactor effluent air fin condenser, as listed in Table 1.
Also, the RPC further enhanced operational reliability by insisting that a layer of macroporous trap/media layer be added above the planned catalyst grading system to capture particulates and iron sulfide that can cause increased reactor pressure drop and thereby shorten the run length. The RPC and the refinery technical service engineer requested use of wedges and pins, instead of traditional nuts and bolts, to facilitate handling of the reactor internals and reduce reactor downtime.
The post-startup audit has revealed that this unit will not constrain refinery operations, and it will be in a position to provide the refinery economic advantage and leverage. Also, an outside review has revealed that this unit has the best performance amongst other similar functional units in the industry.
Experience and practical knowledge results in savings.
Using practical knowledge, the BSR team found opportunities to reduce and increase plant reliability:
Tower trays. After initial work and consultation with tray vendors, the detailed engineering contractor (EC) recommended:
20 new trays for an existing amine contactor when the RPC had expected only minor tray changes
30 new trays for the stripper when the RPC had expected no change at all.
So, the RPC worked directly with the tray vendors with correct information and, as a result, it helped eliminate the need for new trays in both towers and determined that the amine contactor required only a change-out of bolted outlet weirs. This resulted in considerable cost savings and less mechanical work during downtime.
Misdirected fractionation tower feed distributor. Due to increased loads, the stripper tower feed distributor hole area needed to be increased from 18 in.2 to 30 in.2. The tower has single pass trays, and the discharge from the existing feed distributor impinges on a downcomers sacrificial wear plate. The EC designer decided to retain the existing six slots as is and install a second row of slots on the opposite end of the distributor, as shown in Fig. 2.
| Fig. 2. Before and after designs on the |
fractionation tower feed distributor.
With such an arrangement, depending on the new slots angle, the feed stream from the newer slots could impinge on the tray deck below, discharge into the downcomer or impinge on the tower shell. All these scenarios are unacceptable. This design would have led to poor operations and product loss. Typically, sacrificial wear plates and feed diffuser plates are typically installed to break stream momentum. The RPC simply requested that the width of the existing slots be increased from 0.75 in. to 1.25 in., to provide the required additional area; the feed would continue to impinge on existing downcomer wear plate.
Control valve. An inexperienced process engineer would size control valves just for a single case of design rate. In reality, truly seasoned engineers know that a detailed study based on hydraulic models of the piping circuits and associated pump or compressor curves is required to establish control-valve conditions, at not only the design rate, but also at a maximum flow, say at 110% design rate and at a minimum flow say, at 40% design rate. Then the control valve would have different pressure drops at these three flows, as opposed to original data sheets prepared by the EC, which showed constant pressure drops.
Accordingly, the RPC reworked all control-valve data sheets using proper hydraulic models and found that, out of the 39 control valves that were sized, approximately 16 were undersized (of which six were grossly undersized) and 10 were oversized (of which two were grossly oversized). Such improper sizing would have led to control problems and/or would have reduced unit capacity. In the past, the RPC has also seen that an improperly specified control valve can experience severe valve-body erosion, especially for those control valves experiencing cavitation, resulting in unplanned plant downtimes.
Computer aided designs (CADs) are most effective only if backed with proper training and field experience.
Multiple improper pipe routings. Pipes are the arteries in a process unit and must be properly designed for trouble-free operation. Process piping that can experience slug flow and/or have pockets and traps are in a class of critical piping that can cause serious operational and reliability problems and must be properly designed. Slug flow creates unstable operation and/or destructively damaged fittings and vessel inlets and walls from a hammering effect. Slug-flow potential can be managed by:
1) Using a pipe with a smaller diameter, if possible
2) Providing dual risers that can be turned off or on, depending on the flowrates
3) Installing an impingement tee at the end of the risers, where possible.
In low-pressure drop piping, pockets can 1) cause flow restrictions and thus loss of capacity, especially in pump suction piping, and 2) induce slug flow in two phase flow piping. Streamlined piping with shorter pipe runs and fewer fittings, and piping that has strictly downturns past the first starting point riser can reduce operational risk substantially.
Examples from three key piping categories are quoted here to illustrate the critical importance of piping and equipment reviews to guard against problems from slug flows and pockets that can affect plant reliability. In each case, initial CAD drawings incorporated improper piping routings and/or convoluted heart stopping designs.
Overly complex reboiler circulation. Initially, a very convoluted arrangement was proposed as shown in Fig.3-Before view. This arrangement would have resulted in a very poor reboiler operation, potentially leading to upsets and downtime for repairs. The RPC improved reliability of operation by streamlining piping layout, minimizing slug flow potential and, at the same time, reduced large bore piping length from 127 ft. to 76 ft. and fittings from 14 to 10 as shown in Fig. 3-After view.
| Fig. 3. Before and after view of design of the reboilers |
Restriction in pump suction piping. The project required the addition of two new stripper-product pumps next to the existing reboiler circulation pumps with the extension of the existing reboiler pump suction header. New piping is shown with solid lines and existing piping is shown with dashed lines in Fig. 4.
| Fig. 4. Before and after view new product pump layout. |
Initially, the proposed arrangement simply violated the good engineering practice by introducing a vapor pocket in the new suction line shown in the before view. The after view illustrates the design subsequently drawn up by the RPC to improve operational reliability while reducing the new piping length from 60 ft to 53 ft.
Problematic tower feed piping. Fig. 5 shows the initial arrangement proposed by EC and also the final arrangement recommended by the RPC for improved reliability for a two-phase flow service and also lower costs. These benefits were achieved through streamlining piping with fewer fittings downstream of a pressure control valve, eliminating liquid traps and reducing total pipe length from 472 ft to 442 ft.
| Fig. 5. Before and after design of tower feed |
Flexible view on the design issue
Flexible eyes for both the big picture and smaller details help identify missteps and opportunities for greater reliability. Missteps to watch for are:
Incorrect condensate handling, routing and placement. The initial proposal, as shown in Fig. 6-before view, demonstrates multiple engineering missteps:
1) Incorrect design of condensate pots and inlet piping
2) Poor layout of critical condensate piping downstream of the pot level control valves, as this piping can be subject to slug flow and
3) Wrong placement of the final condensate flash drum, D-02.
Condensate pots require bottom inlets and do not require demister pads or anti-vortex baffles.
If the pot design and piping installations had been built per initial proposal, it would have affected unit throughput and reliability. The arrangement then proposed by the RPC is shown in Fig. 6-after view. The RPC also captured the opportunity to place the final flash drum in the corner of the pipe rack leaving the units and thus saved BSR the cost of a total of 400 ft on insulated-traced pipes.
| Fig. 6. Before and after design for condensate |
handling, routing and flash drum.
Fortunately, the early faulty designs introduced in this project by some inexperienced outside engineers were corrected in time to avoid costly rework and unscheduled plant downtimes. There were a number of other equally troublesome missteps in other categories, such as splitter design, heat exchangers and pumps, that were fixed by the RPC. Although only process examples are quoted in this article, the other two members of the refinerys project team supporting the mechanical side and instrument, and electrical had similar experiences requiring corrective actions.
All of the cited examples illuminate current concerns for the oil industry leaders and provide warning signs. A frustrated past colleague who now works for another major US refiner recently said, I have seen some engineering lately that I think a smart 8th grader could do better. In these times, the industry will need to protect itself against such design and engineering faults, missteps and miscalculations that could curtail production, cause unscheduled downtimes and/or escalate project cost into overruns.
The writer went out to speak to three outside engineering directors for feedback on this article. Their input and perspectives and recommendations have been incorporated in the article and its conclusion:
The largest single safeguard for the oil industry and affiliated engineering and technology companies is the retention of a talented team of experienced and versatile in-house engineers who are capable of managing daunting challenges in major projects while capable of new thinking and willing to make appropriate changes to any of the outdated design methods and practices.
Decisions made strictly on the basis of pricing can turn out to be costlier over the longer term, and due diligence is warranted to achieve a proper balance.
Experienced, multi-talented refinery engineers would need to reside and work together with younger staff members and engineering contractors in a truly joint- team fashion and thus successfully meet the projects goals for quality, cost and timeliness and to ultimately capture a savings of 15%30% on engineering costs by eliminating reworks. HP
|The author |
Kirit Sanghavi is a senior refinery process engineering consultant at Alon USAs Big Spring Refinery. He has been with BSR for the past 18 years, responsible for the largest capital projects at the refinery. Previously, he worked at Esso Chemical and Imperial Oil in Canada for 15 years. Mr. Sanghavi has also worked for four international engineering companies in the US, the UK and Canada during his career. He earned a bachelors degree in chemical engineering from London University.