Friday, July 24, 2009

Flowmeter Selection

Flowmeter Selection

When a flowmeter is needed, the selection process should include studying the characteristics of respective measurement technologies and analyzing the advantages/disadvantages for different plant environments. This effort will help ensure that a meter with the right performance and reliability, for a particular installation, is selected. Some of the most common industrial flowmeter designs are described here.

Differential Pressure

A differential pressure meter operates by measuring the pressure differential across the meter and extracting the square root. These meters have a primary element that causes a change in kinetic energy, which in turn creates differential pressure in the pipe. A secondary element measures the differential pressure and provides a signal or readout, which is converted to the actual flow value.

Two basic types of primary elements rely on this measurement: orifice plates and Venturi tubes. Both element types rely on the law of conservation of energy and Bernoulli’s energy equation to determine volumetric flowrates.


Electromagnetic meters (commonly referred to as “mag” meters), employ Faraday’s law of electromagnetic induction, which states that voltage will be induced when a conductor moves through a magnetic field. The liquid serves as the conductor. Energized coils outside the flow tube create the magnetic field. The amount of voltage produced is directly proportional to the flowrate.

Magnetic flowmeters are only applicable for fluids with some electrical conductivity, typically those with conductivity values above 5 µS/cm. Most aqueous solutions contain enough conductive dissolved solids to meet this requirement. However, ultrapure water, some solvents, and most hydrocarbon-based solutions do not.


Coriolis meters provide mass-flow data by measuring fluid running through a bent tube, which is induced to vibrate in an angular, harmonic oscillation. Due to the Coriolis forces, the tube will deform, and an additional vibration component will be added to the oscillation. This causes a phase shift over areas of the tube, and this shift can be measured with sensors. Density measurements are made by analyzing the frequency shift of the vibrating pipe as the fluid flows past the pickup.

Thermal Mass

Thermal mass meters utilize a heated sensing element that is isolated from the path of fluid flow. The flow stream conducts heat from the sensing element, and this heat is directly proportional to the mass flowrate. The meter’s electronics include the flow analyzer, temperature compensator and a signal conditioner that provides a linear output, which is directly proportional to mass flow.

The electrical current required to maintain the temperature at the temperature sensor is proportional to the mass flow through the flowmeter. These flowmeters are commonly used in automobiles to determine the air density as it travels into the engine.

Vortex Shedding

In this instrument, fluid vortices are formed against the meter body. These vortices are produced from the downstream face of the meter in an oscillatory manner. The shedding is sensed using a thermistor, and the frequency of shedding is proportional to volumetric flowrate.


Turbine meters incorporate a freely suspended rotor that is turned by fluid flow through the meter body. Since the flow passage is fixed, the rotor's rotational speed is a true representation of the volumetric flowrate. The rotation produces a train of electrical pulses, which are sensed by an external pickoff and then counted and totalized. The number of pulses counted for a given period of time is directly proportional to flow volume.

Turbine meters are used extensively to measure refined petroleum products, such as gasoline, diesel fuel or kerosene in custody-transfer applications.

Positive Displacement

Positive displacement (PD) meters separate liquid into specific increments. The accumulation of these measured increments over time is given as the flowrate. As the fluid passes through the meter, a pulse, which represents a known volume of fluid, is generated.

Some of the design types included in the positive-displacement flowmeter family include oval gear, rotary piston, helical, nutating disk and diaphragm flowmeters. In all design types, the fluid or gas forces a mechanical element, such as a set of gears, a disk, or a piston, to move within the primary device. For every revolution of a gear, or the complete movement of a piston or plate, a known volume of material is displaced.


Ultrasonic meters operate by comparing the time for an ultrasonic signal to travel with the flow (downstream) against the time for an ultrasonic signal to travel against the flow (upstream). The difference between these transit times is proportional to the flow, and the flowmeter converts this information to flowrate and total flow.

They are particularly useful for measuring the flow of non-conductive fluids, such as solvents and hydrocarbons in large pipes — applications for which a magnetic flowmeter will not work. Ultrasonic flowmeters are also often used in district heating and chilled-water systems.

Doppler ultrasonic flowmeters have one transducer mounted at an oblique angle to the pipe. The transducer generates a signal into the fluid, which is reflected back from suspended particles or air bubbles.

Transit-time ultrasonic flowmeters have two transducers, likewise mounted at an oblique angle to the pipe, on opposite sides of the pipe. Allternating, one transmitter sends sound waves through the fluid to the other.


  1. Keith, J., Evaluating Industrial Flowmeters Chem. Eng., April 2007, pp. 54–59.
  2. Kohlmann, M., Selecting the Right Flowmeter for the Job. Chem. Eng., September 2004, pp. 60–64.
  3. “Perry’s Chemical Engineers’ Handbook,” 8th ed. New York: McGraw Hill, 2008.

Share |

Tuesday, April 21, 2009

HAZID Hazard Identification

HAZID Hazard Identification


A HAZID study is carried out by a team of competent engineers from a mixture of disciplines and is led by a person who is experienced in the HAZID technique. Each area of the installation is considered against a checklist of hazards. Where it is agreed that a hazard exists in a particular area, the risk presented by the hazard is considered, and all possible means of either eliminating the hazard or controlling the risk and/or the necessity for further study are noted on a HAZID worksheet. Actions are assigned to either discipline groups or individuals to ensure the mitigating control, or further study is completed.

  • A hazard can be defined as any operation that could cause an Event (release of toxic, flammable or explosive chemicals, gases or any action) that could result in injury to personnel or harm to the environment.

An operational plant such as a nuclear reactor or critical systems operation such as weapons manufacture, handling and stowage or the operation of a passenger aircraft requires the design of a number of diverse interrelated systems, coexisting in the same limited physical space. The Process of Hazard Identification is the procedure to assess all the hazards that could directly and indirectly affect the safe operation of that plant and or system, and is referred to as the Hazard Identification procedure or HAZID. The procedure of hazard identification is broken down and categorised into the two streams that can affect the system both directly and indirectly, and is referred to as Internal Hazards and External Hazards.

A clear understanding of all the possible event chains leading to the most critical accident scenarios is needed to mitigate for example, the complexity of any plant and its operation, which nevertheless exposes it to any number of accident scenarios from interrelated systems which could provoke failure or domino effect from other systems, components or structures located within the proximity effecting the safe operation of the plant. These hazards are referred to as internal hazards and can include, but are not limited to, radioactive inventory, fires, impacts, overpressures and explosions. Likewise, a screening process is carried out to assess all possible hazards to the safe operation of the plant from all indirect events and can include extreme ambient temperatures, extreme wind, flooding, fire, explosions & overpressures, missiles, toxic gases, seismic, aircraft crash, and electro-magnetic interference, these hazards are referred to as external hazards.

Design basis hazard level

When relating to external and internal hazard assessment, a judgment on the frequencies at which hazard levels should be determined, in terms of Reactor Plant, guidance for this can be found in the HSE NII Safety Assessment Principals paper or (SAPs) and allow for plants that cannot give rise to large radiation doses to be designed against less onerous events. These SAPs therefore require a level of interpretation by the assessor and make it clear that all relevant external hazards should be considered when determining the design basis events for both probabilistic and deterministic safety cases and provide the numerical targets for assessing whether the risk from external hazards is tolerable and ALARP.


Seismic hazard definition should include a reasonable frequency distribution of accelerations, i.e. the ground response spectrum, often called the free field response spectrum. For design purposes, this has been the practice internationally, and in the UK, to use piecewise linear spectra based on a median-plus-standard deviation level of conservatism. Such spectra have been used for the design of new plant and for the design basis assessment of existing plant. More recently uniform hazard spectra (UHS) have also been derived. These spectra were developed for seismic PSA and aimed to have a risk of exceedance uniform for all frequencies (hence their name), unlike the varying conservatism implicit in piecewise linear spectra. UHS spectra have been derived for various confidence levels, including the expected level, which is appropriate for seismic PSA and assessors should consider whether all relevant external hazards are listed in the fault schedule. For natural hazards such a seismic the design basis event should be that which conservatively has a predicted return frequency not exceeding 10-4 per year (often, though not strictly accurately termed as the once in 10,000 year event). For a small proportion of nuclear safety related structures - those with modal frequencies of around 1 Hz or less - it may be necessary to consider long period ground motion arising from a large magnitude distant event. The need arises because the foregoing design spectra’s are dominated by the contribution from small to medium earthquakes with epicentres close to the site and by intent, do not significantly include the separate long period motion. This long period ground motion hazard may be considered separately from the design basis spectra, being a separate, infrequent hazard.

Aircraft crash

For aircraft crash structural demand depends on the mass, rigidity, velocity and engine location of any aircraft assumed to impact directly or skid onto the structure, and also the angle of incidence of the impact (direct or skidding). For these reasons, aircraft are often grouped into a small number of types - eg large commercial aircraft, light aircraft and military aircraft - to facilitate the analysis. In addition to structural effects, fuel fire is highly probable. This will be more significant for the heavier classes of aircraft because of the quantity of fuel carried. It may, however, be possible to exclude some (or all) classes of aircraft on the grounds of low probability (eg well below 10-7 per annum) of impact, thus obviating the need for structural design against impact or fuel fire. In order to assess the probability of impact, the safety case will normally derive an effective "target area" for the site, taking account of the plan area and height of safety related buildings, a representative range of angles of impact and so on, which can then be compared with the aircraft crash frequency per unit area.

Further details of a particular method are published by the IAEA (see below)

The estimated aircraft crash frequency may seek to take into account any flying restrictions which may apply to the site. If so, the assessor should be satisfied that this is justified. Liaison concerning flying restrictions around nuclear licensed sites is handled by NSD's Strategy Unit. The possible effects on safety related equipment from a nearby impact may need consideration.

Where aircraft impact is not excluded in accordance with principle P119, the type or types of aircraft and their associated load/time functions, or a bounding load/time function should be specified. The design basis analysis principles and the PSA principles should be satisfied, as appropriate, taking into account the direct impact of the aircraft on the structures, systems and components important to safety, secondary missiles, vibration effects and the effects of aircraft fuel burning externally to the buildings or other structures, or entering the buildings or structures. Further guidance is available from the IAEA.

Extreme ambient temperatures

The extreme ambient temperature hazard is ameliorated by the slow development of extreme conditions and the relatively long timescales for the plant to respond. It can be assumed that there will be at least several hours notice of extreme conditions developing, and often several days. High temperatures are a potential challenge to electrical equipment which may have essential safety functions. Low temperatures may through brittle fracture of safety related structures and/or the freezing of liquid filled systems pose a threat to safety functions. Low temperatures may also threaten cooling water supplies through freezing. The assessor should establish that the potential threats are recognised by the operators and appropriate prearranged responses are embodied in operating instructions.


Most UK nuclear facilities are potentially subject to flooding both by extreme precipitation directly onto the site, and indirectly from rivers and the sea. As with the other environmental hazards it is important to ensure that the most up-to-date information available for a specific site is used in the hazard assessment. The effects of climate change should also be taken into account (see section 4.5 below). By its very nature the definition of the flooding hazard at small annual probabilities of exceedance will be subject to significant uncertainty and it must be assumed that the natural phenomena which can be the cause of flooding may occur together. For example in the case of sea flooding, extreme wind not only affects wave heights but can also elevate still sea levels through storm surge. Storm surge can be additive or subtractive, and must be combined with the highest and lowest astronomical tides and with barometric effects. The hazard determination should therefore carefully examine the statistical dependencies in combining waves with still sea water levels. The flooding safety case should not be sensitive to the level of the hazard, and operational response may be required. As with the extreme temperature hazard it may be reasonable for the operational response to recognise some warning of extreme flooding, provided the necessary response measures can be initiated with sufficient margin.

Extreme wind

Licensees, any particular application should be assessed to ensure that there are no plant specific, i.e. local aerodynamic, effects which could exacerbate the wind loadings. Typical problems could be wind tunnelling between tall structures, or vortex shedding from upwind facilities. Any structure which is shown to be vulnerable to wind loading should be considered from this point of view and in addition the potential for damage from windborne missiles must be considered. A wind load reduced from one in fifty years to one in two years has been used for the design of some facilities, and is broadly consistent with the foregoing time at risk considerations.

Fire, explosion, missiles, toxic gases

The hazard here will arise either due to the conveyance of hazardous materials on adjacent transport routes (pipeline, rail, road and sea) or adjacent permanent / non-permanent facilities. Typical hazards, which may arise from industrial plants, may be from stored gas, fuel, explosives, pressure vessels or turbine disintegration. The external hazards safety case should consider all potential sources of external missiles and explosion.

Impacts and Shock loading

This can be due to external explosions and overpressures giving rise to shock loading and drop loads or impacts due to facility collapse, which may include cranes, building structures and systems. Here the Licensee should consider the withstand to shock loading, the possible impactors to the system from the facility, determining the largest single mass object from its potential drop height in free fall without structural interaction, or make a case for the probable (but bounding) collapse dynamics of the facility which can in some cases include structural interaction.

Electro-magnetic interference

The potential for electro-magnetic interference to instrumentation and control equipment should be considered. The primary natural source is electrical storms. External man-made sources include radar and communication systems. Depending on whether the hazard can be adequately controlled, the Licensee may need to provide screening within building structures to protect equipment from electro-magnetic interference or install instrumentation and control equipment of a proven electro-magnetic compatibility. Solar flare effects have been known to cause problems on long transmission lines at high latitudes in Canada, but on current knowledge are not expected to cause significant effects at the lower latitudes of the UK with its shorter transmission lines.

Sensitivity studies

It should also be borne in mind that forecast climate change is likely to have an impact on many of the external hazards addressed here. This is likely to include extreme ambient temperatures, wind and flooding. Licensees should be expected to take the latest available predictions over the projected life of the facility, which may need to include the decommissioning phase of the installation in the submissions.

Cliff edge

The Licensee will also need to demonstrate that there will not be a disproportionate increase in risk from an appropriate range of events which are more severe than the design basis event. This is generally known as the cliff edge effect. The way in which this principle is satisfied may depend on the nature of the hazard being addressed. For some hazards a point will be reached where there is a step change in the effect on the installation. In the case of external flooding, for example, the site defences become overtopped. In such cases, it needs to be shown that there is a reasonable margin between the design basis and the point at which this step change would occur. For other hazards, such as seismicity, the forces acting on the installation will continue to increase progressively with increasing size or proximity of the event. A demonstration is needed that there will not be a step change in the response of the installation to the hazard, in terms of the likelihood of a release of radioactivity, for an appropriate range of events more severe than the design basis event. There may be more than one way in which this can be achieved. In the case of seismic engineering, one approach which has been adopted has been to show that the response of the plant remains fully elastic up to a significant margin beyond the design basis. Alternatively, the trend for new design is increasingly to show that the plant will accommodate the seismic forces through a ductile response without any danger of a release of radioactivity occurring. The residual seismic risk from events less probable than the DBE can be a significant contributor to the total risk. It has also been demonstrated in numerous earthquakes that structural ductility is very desirable. Ductility provides a better assurance than elastic margins for the ability to withstand beyond design basis seismic events, and also gives confidence in the ability of structures to cope with the uncertainty in the actual hazard spectrum (peaks etc), uncertainties in the material data, uncertainty in the analyses, and uncertainty concerning other simultaneous loads. Ductility is increasingly being required by nuclear and non-nuclear structural seismic design standards even where the structure is designed to remain elastic under the design earthquake loads. It has previously been accepted that one satisfactory approach to the demonstration of absence of an adverse cliff edge effect is via the PSA. This has the merit, usually, of exploring the response of the plant to a wide range of hazard levels and is accepted internationally as a reasonable approach for external hazards. However, if this approach is adopted, the assessor should ensure that the hazard definition is reasonable for the more remote levels and that relevant equipment responses are reasonable, i.e. important structures are not omitted from consideration by virtue of alternative success paths.

If a PSA is not used to demonstrate the absence of an adverse cliff edge effect either an approximate PSA approach may be undertaken (a NUREG describes a technique for earthquake hazard or a deterministic-plus-engineering judgment approach may be made. As noted above, however, the detail of the approach needs to be appropriate to the nature of the hazard being addressed.

Single failure criterion

Safety systems required in response to any 10-4 annual probability of exceedance external hazard should comply with the single failure criterion. Where this is not feasible in the case of existing facilities, the risk must be shown to be tolerable and ALARP.

Reliability, redundancy, diversity and segregation

In assessing safety systems claimed to mitigate the effects of external hazards, the assessor should have due regard to Reliability, redundancy, diversity and segregation. External hazards may particularly give rise to common mode or common cause failures.

Example HAZID Tables

Share |

Sunday, April 12, 2009

Energy Efficiency in Steam Systems

Energy Efficiency in Steam Systems

In today’s typical process plants, preventing steam loss and improving condensate return are key opportunities to make a process more energy efficient.

To be the most effective, steam generally needs to be dry (such as for process usage), or superheated (for instance, for use in turbines). These requirements dictate utility-system operating procedures for generating the highest quality steam possible, and then distributing it to the points of use with minimal deterioration. Since steam becomes condensate after its heat energy is expended, strategies must be in place to remove condensate as quickly as it is formed, in the steam-supply portion of the circuit and during steam usage alike.

Furthermore, superheated steam is typically desuperheated by injecting hot condensate into the system. As a result, excessive wetness can also occur downstream of the desuperheating station. In either case, if such condensate is not removed from the steam supply, the negative impact on the steam system can be substantial, as seen in Table 1.

Improving condensate return. At many plants, the operators admittedly realize that condensate must be removed as quickly as it is formed, but a suitable condensate drainage or transportation system is not in place. In such cases, the condensate is often sewered or sent to a field drain. Some possible outcomes of removing condensate but not handling it effectively are outlined in Table 2.

Figure 1. Preferred method to drain jacketed pipe for high-melting-point fluids, such as sulfur

Condensate is traditionally removed from steam systems by steam traps or by equipment combinations involving level pots and outlet control valves. Process situations in which high backpressure from the downstream portion of the condensate-return system tend to create a "stall." Then, a different system incorporating both a pump and trap in the design is needed to drive the condensate while also trapping the steam; this process may be referred to as pump-trapping or power-trapping.

Because there are at least three condensate-drainage alternatives, it makes more sense to think in terms of required "condensate discharge locations" rather than referring to condensate removal devices indiscriminately as "steam traps." This broader mind-set helps avoid any predisposition to install steam traps in applications that need a different type of condensate drainage solution.

Figure 2. Alternative, practical redesign method for existing installations to drain jacketed pipe for high-melting-point fluids, such as sulfur (no “stall”)

Engineered separator-drains remove condensate that is entrained in a moving steam supply (including flash or regenerated steam). The result is highest quality steam delivered for plant use. Compare that to steam traps, which remove condensate that has already fallen out of the steam. As their name suggests, steam traps remove condensate and "trap steam." Meanwhile, level pots can be used in certain instances where steam traps cannot meet the high pressure or capacity requirements.

Special situations. There can be many situations in a plant where effective condensate removal requires specialized drainage designs. For instance, Figures 1 and 2 show two options for condensate drainage from a jacketed pipe that conveys high-melting-point materials, such as liquid sulfur or high-boiling hydrocarbons.

Other examples of specialized applications include options to effectively drain steam-supplied heat exchangers. A key consideration is to first determine whether a stall condition exists or not; when it does, condensate will not drain effectively through a simple steam trap. Such a situation typically arises when modulating steam pressure creates a negative pressure differential across the condensate drain device. So-called, Type II secondary pressure drainers of the pump-trap type are used on equipment with a negative pressure differential. Because wasted condensate is a valuable resource to be saved, use Type I secondary pressure drainers of a "pump only" type to recover collected condensate and power it back to the boiler.



Share |

Saturday, February 28, 2009

Flare Gas Recovery

Flare Gas Recovery in Oil and Gas Refineries

O. Zadakbar, A. Vatani and K. Karimpour


Worldwide, final product costs of refinery operations are becoming proportionally more dependent on processing fuel costs, particularly in the current market, where reduced demand results in disruption of the optimum energy network through slack capacity. Therefore, to achieve the most cost-beneficial plant, the recovery of hydrocarbon gases discharged to the flare relief system is vital. Heaters and steam generation fuel provision by flare gas recovery leaves more in fuel processing and thus yield increment. Advantages are also obtained from reduced flaring pollution and extended tip life. During recent years in Iran, all projects have included the collection of associated gases. Thus, flare gas recovery in oil and gas refineries are going to be neglected.

Therefore, in the present work, investigations were made into the operational conditions of 11 important refineries and petrochemical plants. After comprehensive evaluations, we devised practical methods to reduce, recover and reuse flare gases for each petroleum refinery, natural gas refinery and petrochemical plant. The list of refineries and petrochemical plants is shown in Table 1.


Adequate process evaluation of plants, especially the units that produce flare gases, comprehensive monitoring of flow and composition of flare gases, investigation of existing flare systems, and finding alternative choices for reusing flare gases were carried out in 11 petroleum refineries, natural gas refineries and petrochemical plants. The results of the investigation of the existing flare systems, finding alternative choices for reusing flare gases and the overall flare gas recovery system are discussed below.

1.1 Investigation into the Existing Flare Systems

Flare tips are exposed to direct flame during their service life, which can of course be quite damaging. As a result, flare tips need to periodically be taken out of service and refurbished, which adds to production costs. The life of a flare tip is related to the amount of usage. Some of the flares were revealed to burn excess gas due to tip damage. By repairing or replacing these tips, purge gas will be reduced.

Natural gases are typically used as purge gases. This use of natural gases for twenty-four hours each day is not only wasteful of a precious natural resource, it is also very expensive and can represent an expenditure of many tens of thousands of dollars per year. Since air is caused to enter the flare system from the atmosphere only when there is a decrease in the temperature of the gas contained in the pressure-tight flare system, there is a need for purge or sweep gases only when there is a decrease in the temperature of the internal gas content of the flare system. For this reason, there is no need for around-the-clock injection of purge gas for the purpose of avoiding entry of air into the flare system. However, to date, there has been no system for automated injection of purge gases into flare systems only as they are needed, due to gas system temperature decrease. Providing a pair of temperature sensors in the flare gas line is an alternative way to decrease purge gases. These two sensors are placed in close proximity. One is a fast-acting sensor, which responds rapidly to any change in temperature. The other is a slow acting sensor, which responds slowly to a change in temperature. Thus, in combination, they provide a sensor system sensitive to change in temperature in the flare gas line.

To save gas burning, it is recommended to repair or replace pilots with more reliable ones. Also, in some refineries, many ignition systems are fitted, but never in use, because they simply do not work when they are needed. As a result, purge gas is increased to enhance the reliability of the flare. Multi-pilot gas conservation systems are recommended for ignition of waste gas from a flare burner, provisions being made to reduce or limit the pilots for ignition at the most effective location as determined by the wind direction and further, if desired by the wind velocity, a reduction in the number of pilots, effecting substantial savings of combustible gas. A pilotless flare ignitor is an alternative choice too. A pilotless flare ignitor is capable of igniting waste gas issuing continually or sporadically from a flare stack and includes an ignitor housing with an open end which extends into the flare stack.

In some areas a large number of flare stacks have been installed. Global studies recommended optimizing the existing number of flares. In some areas the maximum design capacity of the equipment was reached. Hence, surplus gas is being flared. The studies identified equipment that can be debottlenecked; otherwise, additional new units need to be installed.

1.2 Finding Local Alternative Choices for Reducing and Reusing Flare Gases

Some storage tanks are fixed roof types that require positive pressure set at a certain point. The tank is directly connected to a flare. One of the studies recommended a flow suction tank gas recovery system to be installed at each fixed roof storage tank. The vapor jet system is an alternative to conventional vapor recovery technology for the recovery of hydrocarbon vapors from oil production facilities’ storage tanks. The process utilizes a pump to pressurize a stream of produced water to serve as the operating medium for a jet pump.

In particular, the refinery o.gases from a FCCU contain olefin components, up to about 20 percent by volume ethylene and up to about 11 percent by volume propylene, which components normally are not recovered from the o.gases, but which components may have value to warrant recovery and use in other petrochemical processes or uses in downstream processing.

Delayed coking operations increase the volume of byproduct non-condensable hydrocarbons generated and typically flared. A local flare gas recovery system on a delayed cocker unit is capable of recovering a huge amount of flare gases from the delayed cocker. Using some new environmentally friendly technologies reduces flare emissions and the loss of salable liquid petroleum products to the fuel gas system. New waste

heat refrigeration units are useful for using low temperature waste heat to achieve sub-zero refrigeration temperatures with the capability of dual temperature loads in a refinery setting. These systems are applied to the refinery’s fuel gas makeup streams to condense salable liquid hydrocarbon products .

1.3 Flare Gas Recovery System

Environmental and economic considerations have increased the use of gas recovery systems to reclaim gases from vent header systems for other uses. Typically, the gas is recovered from a vent header feeding a flare. Depending on vent gas composition, the recovered gas may be recycled back into the process for its material value or used as fuel gas. Vent gas recovery systems are commonly used in refineries to recover flammable gas for reuse as fuel for process heaters. The Tabriz petroleum refinery and Shahid Hashemi-Nejad (Khangiran) gas refinery are the most important parts of our work. The results of these case studies are discussed.

1.3.1 Flare Gas Recovery for the Tabriz Petroleum Refinery

The Tabriz petroleum refinery consists of 14 refining units and 10 units related to other services. The nominal capacity of the Tabriz refinery is 80000 barrels per day, but by executing the authorities’ augmenting schemes, nominal capacity has been increased to 115000 barrels per day. The crude oil, up to 115000 barrels in a day, is brought from crude oil preserving tanks to a distillation unit in order to be separated into oil cuts. The necessary crude oil is supplied from the Ahwaz oil fields via a 16-inch pipeline. The Tabriz petroleum refinery normally burns off 630 kg/h gas in flare stacks. The average quantity and quality of flare gas are shown in Table 2.

Having investigated the operational conditions of the Tabriz petroleum refinery, especially the units which produced flare gases, we proposed practical methods to reduce, recover and reuse flare gases for the Tabriz petroleum refinery.

There are some alternative choices for using recovered gases. The most important choices are: using flare gases as fuel gas, for electricity generation and as feed gas. In the next step, we tried to find the best choice for using recovered flare gases. Regarding the operational and economic evaluation, recovery of hydrocarbon gases discharged to the flare relief system is probably the most cost-beneficial plant retrofit available to the refinery. Use of flare gases to provide fuel for process heaters and steam generation leaves more in fuel processing, thus increasing yields. Regarding the results of data analyses, the mean value of molecular weight of the gas is 19.9, and the flow discharge rate is modulated between 0 and a maximum of 800 kg/h. The average temperature is 80C and the average pressure is 1 bar.

1.3.2 Flare Gas Recovery for the Shahid Hashemi-Nejad (Khangiran) Gas Refinery

The Shahid Hashemi-Nejad (Khangiran) is one of the most important gas refineries in Iran. The necessary natural gas is supplied from the Mozdouran gas fields. The Shahid Hashemi-Nejad (Khangiran) Gas Company consists of 5 sour gas refineries, 3 dehydration units, 3 sulfur recovery units, 2 distillation units, 2 stabilizer units and 14 additional units related to other services. The Shahid Hashemi-Nejad (Khangiran) gas refinery normally burns off 25000 m3/h gas in flare stacks. The analysis of operational conditions shows that some units normally produce flare gases more than other units. The compositions of flare gases produced by these units are shown in Table 3. These streams make the main flare stream. In addition, the process specifications of flare gases in the Shahid Hashemi-Nejad (Khangiran) gas refinery are shown in Table 4.

After a comprehensive process evaluation, we devised practical methods to reduce, recover and reuse flare gases for the Shahid Hashemi-Nejad (Khangiran) gas refinery. In addition, the flame igniter system, the flame safeguards and the existing flare tip have to be replaced.

The fuel gas of the Shahid Hashemi-Nejad (Khangiran) gas refinery is supplied by sweet gas treated in the gas treating unit (GTU). Due to a pressure drop in the gas distribution network in Mashhad city in the northeast of Iran, during cold seasons, they encourage using flare gases as an alternative fuel gas resource and eliminating the use of sweet gas produced in a GTU. Regarding the Shahid Hashemi-Nejad (Khangiran) gas refinery recommendations and the operational evaluations, recovery of hydrocarbon gases discharged to the flare relief system is probably the most cost-beneficial plant retrofit available to the refinery.


2.1 Flare Gas Design for the Tabriz Petroleum Refinery

The design considerations include: the flare relief operation and liquid seal drum, the flow and composition of flare gases and the refinery fuel system. The considerations led to a unit design for normal capacity up to 630 kg/h. Our proposed flare gas recovery system is a skid-mounted package which is located downstream of the knockout drum, as all flare gases from various units in the refinery are available at this single point. It is located upstream of the liquid seal drum as pressure control at the suction to the compressor will be maintained precisely, by keeping the height of the water column in the drum. The compressor selection and

design depends on the system capacity and turndown capability. The most appropriate type and number of compressors for the application are selected during the design phase of the project. Liquid ring compressor technology is commonly used because of its rugged construction and resistance to liquid slugs and dirty gas fouling. A number of characteristics which must be taken into account when compressing flare gas are as follows:

The amount of gas is not constant, the composition of the gas varies over a wide range, the gas contains components which condense during compression, and the gas contains corrosive components. A modular design which includes two separate and parallel trains capable of handling various gas loads and compositions is recommended for the Tabriz petroleum refinery.

The recommended system consists of compressors which take suction from the flare gas header upstream of the liquid seal drum, compress the gas and cool it for reuse in the refinery fuel gas system. It includes two LR compressors, two horizontal 3-phase separators, two water coolers, piping and instruments. The compressed gas is routed to the amine treatment system for H2S removal. The e.ect of the devised FGRS on flaring in Tabriz petroleum refinery is shown in Figure 1.

The FGR system with LR compressor for the Tabriz petroleum refinery is shown in Figure 2.

2.2 Flare Gas Design for the Shahid Hashemi-Nejad (Khangiran) Gas Refinery

In this case, the considerations led to a unit design for normal capacity up to 25000 m3/h. The process specifications of the outlet must be similar to refinery fuel gas. The proposed flare gas recovery system is like the proposed system for the Tabriz petroleum refinery. It has a modular design and comprises three separate and parallel trains capable of handling various gas loads and compositions.


The principal potential safety risk involved in integrating a flare gas recovery system is from ingression of air into the flare header, which can be induced by the compressor suction. This could result in a flammable gas mixture being flashed o. inside the system from flare pilots. It should be noted that the FGR unit does not interrupt the flare system and should be able to handle sudden increases in load. Therefore, no modification to the existing flare system will be attempted, but with two exceptions. The connections through which the compressors will take suction on the system, and additional seal drums which will provide extra safety against air leakage into the flare system and allow the buildup of flare header pressure, during compressor shutdown or flare gas overload. Also, the compressor control system does not a.ect the flare system pressure and thus its design will be able to avoid low pressure suction in the flare system during normal operation. When the compressors are not functioning properly, automatic or manual shutdown should result. The flare system will operate as it does now with no compressors. Meanwhile, if the volume of flare gases relieved into the flare system exceeds the capacity of the FGR unit, the excess gases will flow to the flare stack.

If this volume is less than the full capacity of the FGR unit, a spillback valve will divert the discharged gases back to the suction zone to keep the capacity of the flare gas recovery unit constant. Other safeguards to the flare system against air leakage are:

  • the fail-safe shutdown of the FGR unit compressors on low pressure in the flare system.
  • the shutdown of the FGR unit compressors upon high inlet and/or outlet temperatures.
  • adequate purge connections in the downstream of the seal drum.
  • low flow switches in the purge line to the main flare header downstream of the seal drum, to cut in fuel gas as purge gas.


In this section, the results of economic evaluations and the results of emission control are presented. These results were obtained based on 0.11 $/m3 for fuel gas, 6 $/ton for steam and 5 cent/kWh for electricity.

4.1 Economic Evaluations for the Tabriz Petroleum Refinery

The recommended system includes two LR compressors, two horizontal 3-phase separators, two water coolers, piping and instruments. Capital investment to install the FGR system is $0.7 million which, includingmaintenance, amortization and taxes, corresponds to a payback period of approximately 20 months. Another essential e.ect of using the FGRS is gas emission reduction. By using the FGRS in the Tabriz petroleum refinery, we can decrease up to 85% of the gas emission including CO2, CO, NOx, SOx, etc.

4.2 Economic Evaluations for the Shahid Hashemi-Nejad (Khangiran) Gas Refinery

The proposed system for the Shahid Hashemi-Nejad (Khangiran) gas refinery has three LR compressors, three horizontal 3-phase separators, three water coolers, piping and instruments. Capital investment to install the FGR system is $1.4 million, which includes maintenance, amortization and taxes, with a payback period of approximately 4 months. We can decrease up to 70% of the gas emission by using the FGRS in the Shahid Hashemi-Nejad (Khangiran) gas refinery.


It is well known that there are many economical ways to achieve flaring minimization and gas conservation in oil and gas refineries. In order to find these ways, a comprehensive process evaluation of plants, especially units that produce flare gases, comprehensive monitoring of flow and composition of flare gases, investigation of existing flare systems and finding alternative choices for reusing flare gases was carried out in 11 petroleum refineries, natural gas refineries and petrochemical plants. Based on our comprehensive process evaluation, we devised alternatives to reduce gas flaring.

Recovery of hydrocarbon gases discharged to the flare relief system is probably the most cost-beneficial plant retrofit available to the Shahid Hashemi-Nejad (Khangiran) gas refinery and the Tabriz petroleum refinery. Use of flare gas to provide fuel for process heaters and steam generation leaves more in fuel processing, thus increasing yields.

Advantages are also obtained from reduced flaring pollution and extended tip life. In the Tabriz petroleum refinery, 630 kg/h flare gas will be used as fuel gas by $0.7 million capital investment corresponds to a payback period of approximately 20 months, and also 85% of gas emissions will be decreased.

In the Shahid Hashemi-Nejad (Khangiran) gas recovery, 25000 m3/h flare gas will be used as fuel gas by $1.4 million capital investment corresponds to a payback period of approximately 4 months, and 70% of gas emissions will be decreased.

Share |

Thursday, February 26, 2009

Pipe Sizing

Pipe Sizing

Friction Factor

Fluid flowing through pipes experiences resistance due to viscosity, turbulence and roughness of the pipe surface. The Darcy-Weisbach Equation (1) is commonly used for the analysis of steady-state, Newtonian-fluid flow inside pipes. It summarizes the relations between frictional head loss, fluid properties, pipe geometry and discharge.
For laminar flow (Re < 2,100), the friction factor is a function of Reynolds number only.

In turbulent flow (Re > 4,000), f depends upon Reynolds number and pipe roughness.
Hydraulically smooth pipes. In this case, the friction factor is solely a function of Re. For the determination of friction factor, Von Kármán and Prandtl developed Equation (3).
This correlation must be solved by iterative procedures, but simpler correlations given by Colebrook and Blasius are written as Equations (4) and (5), respectively.
Commercial pipe. In this case, f is governed by both Re and relative roughness, expressed as ε / D. The Colebrook-White’s Equation (6) is used to calculate f .
As this equation requires trial-and-error solution, Altshul has developed Equation (7), a computationally simpler choice.

Pressure Drop

To determine pressure drop, discharge and diameter must be known. Hydraulically smooth pipes. Using Equation (1) and the friction factor correlation for smooth pipe, Equation (8) is found.
Commercial pipes. Using Equation (1) and the friction factor correlation for smooth pipe, Equation (9) is found.


To determine discharge, pressure drop and diameter must be known. Hydraulically smooth pipes. Equations (1) and (3) allow us to find an expression for the discharge of a smooth pipe.
Commercial pipes. Equations (1) and (6) allow us to find an expression for the discharge of a commercial pipe.

Pipe Diameter

Rearranging Equation (1) to yield an expression for pipe diameter gives Equation (13).
Smooth pipes. Substituting Equation (5) for f yields a correlation for pipe diameter.
Commercial pipes. Determining the diameter of a rough pipe requires the use of Gu, the dynamic roughness.
Manipulating Equation (7) to reflect Gu and substituting into the expression for pipe diameter gives Equation (17), commercial pipe diameter. Several design parameters can be condensed into a constant, named λ.
The range of Gu is: 0 <>6, based on the known ranges of Re and ε/ D for all pipe and flow conditions. Substituting these two extreme values of Gu into Equation (15) gives the following extreme cases, which a pipe diameter must fall between.
Case 1: Extremely smooth pipe. Gu = 0.
Case 2: Extremely rough pipe. Gu = 10 6
Here, we see that even for very rough pipe (ε/ D = 0.01, Re = 10 8), the diameter estimate will be
only about five thirds of that for smooth pipe.

Graphical Sizing Method

To avoid lengthy calculations, a graphical method can be used to approximate pipe diameter. Dividing Equation (17) by Equation (18), we get the diameter multiplier, Ψ.
A graphical method using Ψ can help to quickly estimate the degree of roughness the chosen pipe can withstand.

Share |

Tuesday, February 24, 2009

Improving Heat Recovery

Compact Heat Exchangers

These units offer distinct advantages over shell-and-tube heat exchangers, as quantified by the example presented here

Johan Gunnarsson Alfa Laval Lund AB, Iain Sinclair and Francisco J. Alanis AspenTech UK Ltd.

Global warming is of major concern today. There is increasing pressure on industry to reduce both energy usage and the associated CO2 emissions. An important and profitable action that industry can take is to recover more process energy and thus improve the efficient use of that energy. This not only reduces the cost of primary energy supply and lowers CO2 emissions, but also provides benefits in terms of reductions in heat rejection and in the associated equipment and operating costs. While making such investments, it is also important that financial returns are maximized and that further opportunities for saving energy and reducing emissions are not missed. This article considers the use of compact heat exchangers (CHEs) for improved heat recovery, as they often achieve higher levels of savings with a better payout rate than more conventional alternatives.

Compact heat exchangers

The dominant type of heat exchanger in process plants today is the shell and tube. In many cases, it is an appropriate selection for the service required. However, because engineers are familiar with shell-and-tube varieties, they tend to select them “by default,” without considering alternatives. If engineers’ minds were opened to alternative technologies, such as compact heat exchangers, many heat-exchanger specifications might look different.

There are many different kinds of compact heat exchangers. The most common is the gasketed plate-and-frame heat exchanger. All CHEs use corrugated plates between the heating and cooling media. The design provides the advantages of high turbulence, high heat-transfer coefficients and high fouling resistance. High heat-transfer coefficients allow smaller heat-transfer areas compared to traditional shell-and-tube heat exchangers used for the same duty. This ultimately results in significant size reductions and weight savings as less material is needed to construct the unit. This is especially important when working with expensive corrosion-resistant metals such as titanium and Hastelloys, for example.

The gasketed plate heat exchanger is often the most efficient solution. In petrochemical and petroleum-refinery applications, however, gaskets frequently cannot be used because aggressive media result in a short lifetime for the gaskets or because a potential risk of leakage is unacceptable. In these cases, all-welded compact heat exchangers without inter-plate gaskets should be considered. There are several different kinds available in the market today. In the case presented in this article, a unit with overall fully counter-current flow is used to enable the required heat recovery, while also allowing mechanical cleaning. In addition, all welds are accessible for repair purposes if this type of maintenance becomes necessary during the life of the exchanger.

When to use CHEs

CHEs can be used in most industrial applications as long as design temperature and pressure are within the accepted range, which normally is up to 450°C and 40 barg. CHEs are often the best alternative when the application allows gasketed or fully welded plate heat exchangers, when a high-grade, expensive construction material is required for the heat exchanger, when plot space is a problem or when enhanced energy recovery is important.

When the application allows shell-and-tube heat exchangers to be manufactured completely of carbon steel, such design normally provides the most cost-efficient solution. However, even in those cases, CHEs can have advantages, such as space savings, superior heat recovery and a higher resistance to fouling, which make them well worth considering.

If you do not know if your application can be handled by compact heat exchangers, ask a vendor. Suppliers are normally willing to give you a quick budget quote when their equipment is appropriate for your application so that you can compare solutions and determine which would be best for you. As part of the vendor enquiry, design options for enhanced heat recovery can be quantified and additional energy saving benefits and capital cost changes can be defined. At this stage, in some circumstances, it may be favorable to respecify the heat-exchanger performance requirements to take advantage of the improved heat recovery that can be achieved with a CHE.

CHE versus shell-and-tube

All-welded CHEs consist of plates that are welded together (Figure 1). Among the many models available on the market today, all have one thing in common: they do not have inter-plate gaskets. This feature is what makes them suitable for processes involving aggressive media or high temperatures where gaskets cannot be used. Figure 1. All-welded compact heat exchangers are very compact compared to shell-and-tube heat exchangers.

On the other hand, some of these all-welded heat exchangers are sealed and cannot be opened for inspection and mechanical cleaning. Others can be opened, allowing the entire heat-transfer area and all welds to be reached, cleaned and repaired if necessary.

Because all-welded heat-exchanger plates cannot be pressed in carbon steel, plate packs are available only in stainless steel or higher-grade metals. The cost of an all-welded compact heat exchanger is higher than that of a gasketed plate heat exchanger. Nevertheless, in cases where gaskets cannot be used, all-welded compact plate heat exchangers are still often a strong alternative to shell-and-tube heat exchangers.

The most-efficient, compact, plate-heat-exchanger designs have counter-current flows or an “overall counter-current flow” created by multi-pass arrangements on both the hot and cold sides. Such units can be designed to work with crossing temperatures and with temperature approaches (the difference between the outlet temperature of one stream and the inlet temperature of the other stream) as close as 3°C.

As mentioned before, all-welded CHEs are very compact in comparison to shell-and-tube heat exchangers. CHEs have this advantage due to their higher heat-transfer coefficient and the resulting much smaller heat-transfer area. The units typically occupy only a fraction of the space needed for a shell-and-tube exchanger. Space savings are accompanied by savings on foundations and constructional steel work, and so on. The space needed for maintenance is also much smaller as no tube-bundle access and withdrawal space is required.

Due to the short path through the heat exchanger, the pressure drop can be kept relatively low, although this depends on the number of passes and the phase of the fluid. For most liquid-to-liquid duties, a 70 – 100 kPa pressure drop is normal, while for a two-phase flow, the pressure drop can be as low as 2 – 5 kPa.

Regarding heat recovery, the main advantage of the CHE is that it operates efficiently with crossing temperatures and close temperature approaches. This makes it possible to transfer more heat from one stream to another or to use a heating medium that is just a few degrees warmer than the cold medium.

There are two main reasons why all-welded CHEs are more thermally efficient than shell-and-tube heat exchangers:

  • All-welded CHEs have high heat-transfer coefficients. This is due to the high turbulence created in the corrugated plate channels. The high turbulence results in thin laminar films on the surface of the heat-transfer area. These have a much lower resistance to heat transfer compared to the thicker film found in a shell-and-tube heat exchanger
  • Counter-current flows (or overall counter-current flows) can be achieved in all-welded compact heat exchangers. This means that a single heat exchanger, operating with crossing temperatures and a close temperature approach can replace several shell-and-tube heat exchangers placed in a serial one-pass arrangement, to emulate the counter-current flow of the compact heat exchanger design

As a result, CHEs may be more cost-effective and may present a more practical alternative to shell-and-tube heat exchangers. In addition to the financial benefits, space savings can also be an important factor for upgrading existing plants as well as for new plant designs.

The advantages of CHEs over shell-and-tube heat exchangers will become clear with the following example taken from an actual application.

A real application example

In a recent feasibility study for improving the energy efficiency of a European ethylene plant, a number of opportunities to increase the export of high-pressure (HP) steam to the site’s utility system were identified. The changes included unloading the refrigerant compressors and increasing heat recovery from the quench water loop.

One such opportunity was the replacement of an existing quench water/polished water shell-and-tube heat exchanger that was limiting heat recovery. From an energy point of view, it was desirable to maximize heat transfer between these streams. This would reduce both the low-pressure (LP) steam required for boiler feed water (BFW) deaeration (due to an increase in deaerator BFW feed temperature) and would also reduce the heat-duty load on the cooling water tower (a site bottleneck), due to a reduction in quench water cooling against cooling water.

The required minimum performance of the replacement heat exchanger is detailed in Table 1.

A preliminary assessment of the suitability of a shell-and-tube heat exchanger indicated that two shells in series (468 m2) would be an economical compromise, achieving a heat recovery of 10 MW with an 11.6˚C temperature approach at the hot end.

At this stage, a compact heat exchanger was compared with the shell-and-tube alternative. An all-welded rather than a gasketed plate heat exchanger was chosen because of limited gasket lifetime when there is contact with quench water. Additionally, because of potential quench-water side fouling, an all-welded heat exchanger that could be mechanically cleaned was preferred.

As mentioned previously, selecting an all-welded CHE instead of a shell-and-tube heat exchanger makes it possible to further increase energy savings, by reducing temperature approach. In this case, the hot-end temperature approach determines the duty and thus the size and design of the heat exchanger. For a compact heat exchanger with counter-current flows it is normally possible (and economical) to decrease the temperature approach to 3 – 5°C. To take advantage of this potential, various improved heat recovery designs were investigated.

A summary of alternative heat-exchanger designs is shown in Table 2. There, it can be seen that the heat-transfer coefficient for the compact heat exchanger is much higher than for the shell-and-tube heat exchanger. This is due to the highly turbulent flow created by the corrugated plates in the CHE. As a result, a much smaller heat-transfer area is required. When comparing the cost of the all-welded CHE and the shell-and-tube heat exchanger, it should be remembered that the plate material in the CHE is stainless steel (ANSI 316L), while carbon steel is used in the shell-and-tube heat exchanger.

It should also be noted that the pressure drop is higher for the compact heat exchanger than for the shell-and-tube heat exchanger. This will, of course, increase the fluid-pumping cost. A true comparison must take these costs into account. However, since the pumping costs are usually small when compared to the overall energy savings achieved, the financial outcome for this example is unlikely to change.

The installation cost of shell-and-tube heat exchangers will be higher, especially for a multi-shell design. In this case, the total installed cost comparison would therefore be significantly more favorable for compact heat exchangers than the purchase cost comparison given above.

For the heat exchangers considered in this example, Table 3 shows how energy and emissions reductions improve as the cold-side outlet temperature is increased to reduce the hot-end temperature approach from 11.6°C to 3.9˚C. To achieve this, 50% more compact heat exchanger surface area is required. This increases the cost of the unit by only 26%; however, on the other hand, two shell-and-tube heat exchangers in series would be required to achieve the same performance, which would require 85% more heat-transfer area, at a 69% higher cost.

All design options offer reasonable monetary savings. Heat exchanger selection is therefore primarily driven by capital cost. A compact heat exchanger design allows improved heat recovery with only a marginally longer payback time, and therefore, is a strong candidate for selection.

Figure 2. For a compact heat exchanger with counter-current flow, as shown in Figure 3, it is normally possible to decrease the temperature approach to 3–5°C.

Figure 3. Counter-current flows can be achieved in all-welded compact heat exchangers. This means that a single heat exchanger, operating with crossing temperatures and close temperature approach, can replace several shell-and-tube heat exchangers placed in a serial, one-pass arrangement.

The all-welded compact heat exchanger in Case 3a provides maximum energy savings and CO2 credits at a lower size, cost and payback time than the corresponding shell-and-tube heat exchanger in Case 3b. With 17% additional monetary saving, the payback time for the compact heat exchanger is only 8% longer, whilst the payback time for the shell-and – tube heat exchanger design is 44% longer.

The following two points should also be noted:

· The installation cost of the all-welded CHE should be lower than for a shell-and-tube, especially when the shell-and-tube design is a multi-shell arrangement, as in this comparrison

· All-welded CHEs often provide better lifecycle performance and lower maintenance costs than shell-and-tube designs, because there is less fouling. Less fouling means less-frequent cleaning, which in turn reduces downtime (or at least the maintenance work). Compact all-welded heat exchangers are also very easy to clean. Their panels can simply be removed to allow mechanical cleaning with high-pressure water. Shell-and-tube heat exchangers, on the other hand, take longer to clean

Final remarks

There is increasing pressure on industry today to reduce CO2 emissions. Reducing energy use by improving process heat recovery, is an effective way for companies to respond to this pressure.

Reducing energy use lowers costs for primary energy supply and thus reduces operating costs. Also if primary energy supply is reduced, heat rejection must also reduce. Overall, the capital investment cost for all heat transfer equipment is often lower.

It is our experience that opportunities for improved heat recovery and reduced CO2 emissions exist in most chemical process industries (CPI) plants, and that some of these opportunities can be realized with short payback times. This allows companies to contribute to CO2 reduction initiatives and to reap financial benefits.

Effective feasibility studies for reducing energy use should follow a systematic approach and involve equipment vendors, to ensure that all potential opportunities are fully exploited.

Finally, all-welded compact heat exchangers can often improve heat recovery, while achieving greater savings with a better payback rate than more conventional alternatives such as shell-and-tube heat exchangers.

Share |


Twitter Delicious Facebook Digg Stumbleupon Favorites