emerson bypas sag365652

8/20/2019 Emerson Bypas Sag365652

http://slidepdf.com/reader/full/emerson-bypas-sag365652 1/4

28 | Valve M A G A Z I N E

DESUPERHEATING FOR

ACCURATE STEAMTEMPERATURE CONTROL

THE SEEMINGLY SIMPLE PRACTICE OFADDINGWATERTO STEAMTO LOWERITSTEMPERATURE IS ACTUALLY QUITECOMPLEX.THE AUTHORPROVIDESGUIDELINESTO HELP OPERATORSUNDERSTANDHOWTO CONTROLTHEDESUPERHEATINGPROCESS.

BY JOEL W. KUNKLER

Temperature SetpointThe downstream temperature setpoint should not be too close to the saturationtemperature of the primary steam. This is because as the saturation temperatureis approached, the steam flow almost always exhibits two-phase characteristics.Injection of spraywater, especially in larger pipelines, can result in uneven distri-bution of the steam’s temperature. For example, if the desuperheating spraywaterhas not been properly injected, regions of superheated steam can surround a coreof much cooler steam.

This situation is compounded when the setpoint is near saturation. If some

steam flow is converting to water, droplets will cling to the temperature-sensingelement as the hotter steam passes. This results not only in a false temperaturereading of the steam saturation, but also a cycling of the desuperheating system.The temperature controller reading will increase and the controller will decreasethe spraywater flow continuously while hunting for the correct temperature.

The general rule is that you should have a setpoint greater than 10° F abovesaturation when using feedback control based on a downstream temperature sen-sor. If a setpoint of less than 10° F is absolutely necessary, a feed-forward controlstrategy must be used. This requires a simple algorithm in the plant’s distributedcontrol system to calculate the spraywater required to reach the temperatureneeded based on the conditions of entering steam and spraywater. Also note that asufficient pipe drain system should be part of any desuperheating station to pro-

tect against unexpected overspraying or water fallout situations.

Spraywater PressureThe amount of pressure differential between the spraywater and the steam is veryimportant for both water atomization and the rangeability between maximum andminimum water flows. The maximum pressure differential, along with spraywatertemperature and spray nozzle design, directly affects atomization to the smallestdroplet size: the smaller the droplet, the more rapid the vaporization. Additionally,the greater the pressure differential, the greater the spray nozzle’s rangeability toreach lower water flow situations through continued acceptable differential levels.

Spraywater pressure ideally is 150 to 1,000 psid greater than the steam pres-sure. Although desuperheating devices can operate at much lower differentials, a

In the process and power indus-tries, steam is used both to per-

form mechanical work, such asdriving a turbine, and as a heattransfer fluid. Unfortunately, bothfunctions are accomplished bestwith steam properties at oppositeends of a spectrum—dry super-heated steam is at the high end,while desuperheated steam near itssaturation point is at the low end.Going from the high end of thespectrum to the low end involvessteam conditioning, a process typi-cally misunderstood and oftenoverlooked.

What complicates steam condi-tioning is temperature control, ordesuperheating. This seeminglysimple practice of adding water tosteam to lower its temperature isactually quite complex becausedesuperheating leads to a tempo-

rary, two-stage, liquid-vapor flowwith potential control difficulties.

The three general process rea-sons for desuperheating steam are:

to improve thermal efficiencyof heat transfer processes byusing steam near saturation

to control unintentionalsuperheat from reducing thepressure of steam

to protect downstream equip-ment and piping from elevat-

ed temperatures and pres-sures.

The goal of steam desuperheat-ing is to reduce the temperaturesetpoint at the shortest possiblepiping distance and elapsed timewhile avoiding damage from two-phase flow. A number of criticalinstallation and application param-eters influence whether or not thisgoal is reached.

© 2006 Valve Manufacturers Association. Reprinted with permission.

AS SEEN IN THE FALL 2006 ISSUE OF...



F a l l 2 0 0 6 | 29

direct correlation exists between dif-ferentials and vaporization speed per-formance as well as ability to obtaincontrollable low-flow levels. One cau-tionary note: when considering highpressure differentials, the spraywatercontrol valve may need cavitation pro-tection.

Spraywater TemperatureThe temperature of the spraywater iscritical to rapid vaporization and con-

version into steam. Hotter watervaporizes faster than cooler water fortwo reasons. First, the hotter water iscloser to its saturation temperature soit requires less heat input from sur-rounding steam, and therefore, lesstime to vaporize. Note that an increasein the amount of “hot” spraywater isrequired compared to “cold” waterand the reduction in evaporation ismore favorable than water flowincrease.

The second, more subtle reason forusing hotter water is that at highertemperatures, hot water atomizes intosmaller droplets due to less surfacetension. As a rule, water is deemed hot

at approximately 180° F and greater.The higher above 180° F, the better theatomization.

Note also with caution that poten-tial flashing issues exist both in thespraywater control valve and at thenozzle. Flashing of the spray water asit exits the nozzle is beneficial. Howev-er, flashing upstream, in either thevalve or just before the nozzle, drasti-cally inhibits performance and maydamage both pieces of equipment.

Initial and Final SteamSuperheatThe initial amount of superheat reduc-tion needed determines the amount ofspraywater flow. The greater theamount of spraywater, the longer ittakes for complete vaporization.Equally important, however, is the con-verse: the desired amount of finalsuperheat. As mentioned above, con-trolling to a setpoint barely above satu-ration makes the vaporization process

more difficult.

Minimum Steam VelocityOne of the most critical aspects ofwater vaporization involves minimumsteam velocity. For vaporization tooccur, water droplets must remain sus-pended in the steam flow until they cancompletely evaporate.

The type of spray nozzle and the tur-bulence of the steam flow influenceswhat velocity is required. In a tradi-

tional desuperheater-only configura-tion, the minimum steam velocity mustbe approximately 30 feet per second orgreater. However, lower velocities ofapproximately 10 feet per second arepossible when special desuperheaterconstructions assist in mixing.

The optimum situation, however, isto have steam pressure reductionoccurring immediately upstream of thedesuperheater. Such a situation occurswhen a combined function device—

such as a steam conditioning or turbinebypass valve or a separate pressure-reducing valve located within approxi-mately three to five pipe diameters—isused. Either arrangement can keep

water droplets suspended in averagevelocities as low as approximately fivefeet per second because of the turbu-lence in the steam flow.

Maximum Steam VelocityConcern about the effect steamvelocity has on desuperheating comesfrom the fact that the faster thevelocity, the faster two-phase flowmoves in the pipe and the greater thedistance required to completelyconvert the flow to steam. Highvelocity can be beneficial because itsgreater overall turbulence enhancesthe mixing. However, weight the valueturbulence against the sheermomentum of the steam, which causeslonger distances/time for thespraywater to vaporize. Most steampiping velocity guidelines suggest amaximum velocity of 200 to 250 feetper second to minimize turbulence-induced vibration.

Pipeline SizeAlso important is the size of thepipeline in relationship to the amountof required spraywater. Large amountsof spraywater in small pipelines canlead to water impingement on the pipewall and subsequent fallout. Still, desu-perheating steam in large pipelines canbe challenging because establishing ahomogeneous mixture of steam andinjected water is difficult. This mixing

Figure 1—

This steam

conditioning

valve assembly

features an

integral

desuperheater

section.

Figure 2—A desuperheater that injects

spraywater in the outlet of its venture section

assures excellent mixing and rapid

atomization.




challenge leads to inaccurate tempera-ture measurements and subsequently,poor temperature control.

Installation OrientationOrientation of the desuperheater also

can affect the speed of vaporization.Installations in which spraywater isinjected into a horizontal pipe are mostcommon and are used as the baseline.Installations in a vertical flow-up pipeperform slightly better because of thepositive effect gravity has on the inject-ed water droplets—a longer residencetime enhances vaporization. Converse-ly, however, installations in a verticalflow-down pipe perform slightly worsethan horizontal because of the negativeeffect of gravity—reducing residencetime.

TurndownTurndown is often misunderstood. To auser, the term “turndown” refers to theratio of the maximum to minimumsteam flow a desuperheater can con-trol. However, this view fails to recog-nize the importance of variations inboth steam and spraywater pressures,temperatures and flows that occur atvarious operating times.

The correct method for determining

turndown is to view it in terms of thespraywater nozzle’s control rangeabili-ty. The driving consideration for tem-perature control is the nozzle’s abilityto create an adequately formed, coni-cal-shaped spray pattern. In turn, that

pattern must be comprised of dropletsize and shape easily converted andmaintained over a range of conditions.The spray nozzle’s ability to performdefines the range between controllablemaximum and minimum spraywaterflow, which in turn determines whatcan be accomplished given a certain setof conditions.

StrainersIf the spraywater may include particu-lates such as weld slag, dirt or otherdebris or if the spray nozzle has a verysmall orifice, the use of in-line spray-water strainers is mandatory. Thesestrainers protect the spray nozzle frombecoming clogged by debris, which candecrease capacity as well as distort thespray pattern droplet size.

Steam Pipe LinerLiners are used to protect the steampipes against water impingement andthermal shock at the point where

spraywater is injected. These pipe lin-ers are usually high-grade chrome-molybdenum, which has a greaterresistance to cyclic thermal stress thancarbon or lower alloy steels.

In the past, liners were commonlyused because nozzle design technologywas not as sophisticated and becausewe had less understanding of the ther-modynamic issues involved in desuper-heating. A problem with liners, howev-er, is that if spraywater comes in

contact with them, they ultimately dis-integrate and are released downstreamwith the potential for serious damage.

Often, more careful consideration ofinstallation factors can replace theneed for such a device. However, whenno alternative is available and thepotential for spraywater fallout is toogreat, a liner can improve system per-formance significantly as well as pro-tect against cracking of the main pres-sure-retaining pipes. However, the

increased velocity must be factored

30 | Valve M A G A Z I N E

D E S U P E R H E A T I N G F O R A C C U R A T E S T E A M T E M P E R A T U R E C O N T R O L

Figure 3—This desuperheater has multiple,

fixed-geometry spray nozzles for applications

with nearly constant loads.

Figure 4—Types of desuperheating devices




F a l l 2 0 0 6 | 31

into the calculation for straight pipedistance and temperature sensor loca-tion. The desuperheater supplier mustknow liner dimensions to avoid instal-lation problems. Also, the liner’s con-dition should be checked regularly.

Straight Pipe RunRequirementThe correct distance for uninterruptedpipe after a desuperheater can be deter-mined only after careful considerationof all influencing factors. This distanceshould not contain elbows, valves orother steam flow obstructions. If thestraight pipe is not long enough, non-vaporized water droplets will contact

the first elbow pipe wall and fall out ofthe steam. This unevaporated water willlessen the effect of the injected spraywa-ter, resulting in higher temperatures andrequiring the addition of more water.This will lead to spraywater falling outof suspension. Furthermore, the portionof the elbow pipe wall hit by the spray-water will erode.

In addition to no obstructions in thestraight pipe distance, there should beno piping tees or branches. This isbecause the water in the steam flowcannot be divided equally between mul-tiple flow paths before the temperaturesensor. Modeling, testing and fieldexperience suggest that at least 80 to90% of spraywater droplets should beevaporated before the first obstruction.After that point, elbows tend toenhance the vaporization of theremaining droplets through increasingturbulence in the flow stream.

Distance to TemperatureSensorEqually important is how far the point ofspraywater injection is from the locationof the temperature-sensing device. Theremaining 10 to 20% suspended spray-water must be vaporized completelybefore it encounters this sensor. Waterthat remains in suspension causes inaccu-rate temperature readings from dropletsthat form on the sensor.

Beyond the GuidelinesIt’s important that all the factors thatimpact steam conditioning be consid-ered before applying the separateguidelines. For example, some recom-mendations are made based almost

solely on velocity such as the fact thathigher steam velocity causes greaterturbulence, faster mixing and requiresshorter downstream distances. Whilegreater turbulence is beneficial, theresidence time required by the waterdroplets cannot be ignored.

Water droplet size is sometimespromoted as an absolute criterion. Infact, spraywater nozzle technology hasevolved to the point that some nozzlemanufacturers publish droplet size.However, droplet size is also influencedby spray water pressure and tempera-ture (e.g., hotter water at larger pres-sure drops provides smaller dropletsthan cooler water without much pres-sure drop.)

Instead of relying on general rulesof thumb, it’s best to use a predictionmethod that takes into account all fac-tors, including water temperature,steam velocity, percentage of spraywa-ter and the amount of superheat indetermining straight pipe run length

and sensor location. VM

JOEL W.K UNKLER is a senior applications spe-

cialist in the Fisher products Severe Service

Group at Emerson Process Management in

Marshalltown, IA. He joined the organization

in 1994 and has more than 28 years

experience in the process control industry.

Reach him at 641.754.2533 or at

[email protected].

SPL

Figure 5 – Straight Pipe Length (SPL): assumes 80% vaporization and thermal mixing; calculated

by thermodynamics of system; and without obstruction typically over the first 10 to 40 feet.

TSL

TE

Figure 6 – Distance to Temperature Sensor (TSL): assumes complete vaporization and thermal

mixing;calculated by thermodynamics of system; and very condition specific—can vary from 40

to 100 feet depending on mass and velocity considerations.


emerson bypas sag365652

Documents