eyectors.pdf

8/10/2019 eyectors.pdf

http://slidepdf.com/reader/full/eyectorspdf 1/110

Steam Jet Ejectors in Pilot and Production PlantsSince pi lot plant work is on a small scale, the types of steam jet ejectorsthat are used are not necessarily scaled-down versions of productionplant ejectors. Economy of steam and water is not the governi ng factor;fi rst cost is of more interest.

W. D. MAINS R. E. RICHENBERG

GRAHAM MANUFACTURING CO., INC., GREAT NECK, NY GRAHAM MANUFACTURING CO., INC., BATAVIA, NY

STEAM JET EJECTORS ARE EMPLOYED in the chemicalprocess industries and refineries in numerous and very often

unusual ways. They provide, in most cases, the best way to pro-duce a vacuum in these process plants because they are ruggedand of simple construction—therefore, easily maintained. Theircapacities can be varied from the very smallest to enormous quan-tities. Because of their simplicity and the manner of theirconstruction, difficulties are unusual under the most extreme con-ditions. They are simple to operate. Ejectors which are properly

designed for a given situation are very forgiving of errors in esti-mated quantities to be handled and of upsets in operation and arefound to be easily changed to give the exact results required.

In pilot plant operations all of these are important functions,because in a pilot plant a great deal of information is usuallyunknown, and something must be selected which will operateover a very wide range.

Therefore, this article will outline the differences between ejectorsfor a pilot plant and those for a production plant, pointing outthat pilot plant ejectors are not just small editions of productionplant ejectors.

In order to become fully versed in the essential elements thatmake up a steam jet ejector, the principle of operation will beconsidered first.

In reference to Figure 1, there are the following parts:

1. The steam chest through which the propelling steam isadmitted

2. The steam nozzle through which the propelling steamexpands and converts its pressure energy into kinetic energy

3. The air chamber through which the air, gas, or vapor to beevacuated enters and distributes itself around the steam noz-

zle

4. The diffuser through which the steam and entrained load iscompressed and discharged at some pressure higher than thesuction

All steam ejectors, no matter how many stages and whether theyare condensing or noncondensing, operate on this principle, eachstage being another compressor.

TYPES OF EJECTORS

Ejectors may, in one sense, be put into two categories: condensingand noncondensing. Figure 2 illustrates a three-stage condensingair ejector and a three-stage noncondensing air ejector.

The condensing type utilizes condensers between ejector stages toremove condensable vapor and, therefore, require a source of cooling water. The noncondensing ejector has its stages connecteddirectly together, with succeeding stages handling the motivesteam from preceding stages. This type requires no cooling water.However, it uses considerably more steam than the condensingtype to handle a given load.

Ejectors for pilot plants differ from production plant units inas-

much as noncondensing~type units are normally recommendedwherever possible. However, they may be, and sometimes are,similar. Of course, pilot plant ejectors are smaller, since they aredesigned to handle a smaller load than handled by full-size pro-duction plant ejectors.

In Figure 3, note the plot of ejectors which have been selected forstandard-size plants. To make up this family of curves, steam andwater consumption have been considered, and the ejector type shownis one that has a reasonable steam consumption with a reasonablefirst cost. Usually, higher first cost means lower steam consumptionand better economy in the long run.

Chemical Engineering Process, M arch 1967 1

Figure 1. Conversion of steam pressure into veloci ty in the steam

nozzle and conversion of velocity into pressure in the diffuser.



Schutte & Koerting • 2510 Metropolitan Drive • Trevose, PA 19053 • USA • tel: (215) 639-0900 • fax: (215) 639-1597 • www.s-k.com • [email protected]

Steam Jet Ejectors

Bulletin 5E-H

Introduction

Schutte & Koerting has a century of experience in

designing and building efficient jet vacuum ejectors. This

vast experience allows S & K to handle virtually any jet

ejector application—no matter how complex.

Steam Jet Ejectors are based on the ejector-venturi

principle. In operation, steam issuing through an

expanding nozzle has its pressure energy converted to

velocity energy. A vacuum is created, air or gas is

entrained and the mixture of gas and steam enters the

venturi diffuser where its velocity energy is converted intopressure sufficient to discharge against a predetermined

back pressure.

Jet vacuum ejectors are readily available in ductile iron,

steel, stainless steel and, on special order, in many more

Index

Description Page

Introduction 1

Advantages 2

Performance Characteristics 3

SINGLE STAGE EJECTORS

Fig. 556 (Standard Construction) 4

Fig. 555H (Haveg) 5

Fig. 562/555G (Graphite) 6

Fig. 557/542 (Flanged) 7

MULTI-STAGE EJECTORS

Condensing and Non-Condensing Types 9

Two Stage 10

Three Stage 11

Four, Five, and Six Stage 12

Ejectors with Surface Condensers 13

Non-Condensing Types 13

Low Level Vacuum Units 14

Corrosion Resistant Units 14

Vacuum Boosters 15

Application Considerations 16

Measurement of Low Absolute Pressures 16

Quotation Information 17

Applications 18

materials such as Monel, Alloy 20, Hastelloy, Silicon

Carbide, Titanium, Bronze and others. They can also be

made from a variety of nonmetals such as Haveg,

Graphite and Teflon.

Steam jet ejectors are used in the process, food, steel

and allied industries in connection with such operations

as filtration, distillation, absorption, mixing, vacuum

packaging, freeze drying, dehydrating and degassing.

They will handle both condensable and non-condensable

gases and vapors as well as mixtures of the two. Small

amounts of solids or liquids will not cause operating

problems. Accidental entrainment of liquid slugs can

cause momentary interruption in pumping, but no

damage to equipment.

All S & K ejectors are computer designed and type-tested

to insure reliability.




Steam Jet Ejectors Bulletin 5E-H

Advantages

The principal advantages of steam jet ejectors over other

types of vacuum producing units are...

LOW COST. Pumps of the ejector type are small in relation

to the work they do and their cost is low in comparison

with other types of equipment.

No MOVING PARTS. These units have no moving parts to

adjust or repair.

SIMPLE, COMPACT CONSTRUCTION. Nothing could be

simpler than a jet vacuum ejector. It consists of an

expanding nozzle, a body, and a venturi (or diffuser).

RELIABILITY. Because of their inherent simplicity, these

pumps are reliable. Maintenance requirements are

simple and are easily accomplished.

CORROSION /EROSION RESISTANCE. Units can be made in

practically any workable material to provide utmost

resistance to corrosion and erosion. Standard models are

supplied in a choice of materials as indicated in this

bulletin.

EASY INSTALLATION. Relatively light in weight, jet ejectors

are easy to install, require no foundations. Even multi-

stage units are readily adaptable to existing conditions.

HIGH VACUUM PERFORMANCE. Steam jet ejectors can

handle air or other gases at suction pressures as low as

three microns Hg. abs.




Bulletin 5E-HSteam Jet Ejectors

Fig. 3. Suction Pressure Ranges of Single and Multi-Stage Steam Jet Ejectors

Fig. 4. Suction Pressure Ranges of Single-Stage Ejectors.

Performance Characteristics

The graph, Fig. 3, shows the relative suction

pressure capabilities of S & K Steam Jet

Ejectors from single-stage through six-stage

types. It can be seen that in some cases unitsoverlap. When this occurs, a detailed

comparison of initial costs and steam

consumption should be made before making a

decision as to the exact type required to meet

specific requirements. S & K engineers should

be consulted for their recommendations based

on experience in many applications. Single-

Stage Ejectors are made in several models to

meet various suction pressure requirements.

Fig. 4 shows the range of suction pressure

offered by each model.

A feature of the standard S & K line is that userscan select a size ideally suited for individual

requirements. In addition, a new and carefully

tested design provides far greater capacities

than ever before available. The smallest size

unit now covers a range that previously required

two ejectors of earlier design.

98

76

54

3

2

9

8 76

54

3

2

98

76

54

3

2

98

76

54

3

2

98

76

54

3

2

98

76

54

3

2

1000 MM

100 MM

100 MICRON

10 MICRON

1 MICRON

10 MM

1 MM

1 2 3 4 5 6

810 M Normal Design Discharge Pressure

Lower End Indicates

Zero Suction Capacity

Stages

S u c t i o n P r e s s u r e H g . A b s .

O N E

T W O

T H R E E

F I V E

F O U R

S I X

Relative Suction Capacity - PPH Dry Air

S u c t i o n P r e s s u r e - I n . H g . A b s .

01 1000

30

S-8

S-7

S-6

S-5

S-4

S-3

S-2

S-1





SINGLE-STAGE EJECTORS

FIG. 556 STANDARD CONSTRUCTION

Application

S & K Single Stage Ejectors are designed to cover a suction pressurerange from 1" to 30 " Hg Absolute utilizing eight specific internals as

shown in Fig. 4, page 3 and are used in application of the type noted on

page 4.

Each of the "S" types indicated will produce the most economical

performance in its specific suction pressure range.

Construction

The standard Fig. 556 Single Stage Ejector comprises a converging-

diverging steam nozzle, a body or suction chamber, and a venturi tail

(diffuser).

Sizes 1" through 4" are cast in ductile iron with Type 316 stainless steel

Fig. 5. Fig. 556 Steam Jet Ejector.

Fig. 6. Fig. 556 Steam jet Ejector. This

design is standard for models with

1”, 1 1/2”, 2”, 2 1/2”, 3” and 4” suction

connections.

Fig. 7. Fig. 556 design for models

with 5” and 6” suction connections.

Sizein

Inches

Unit Dimensions Connections NetWeight

(Lbs.)A B C D E F G

1 11 19/64 8 7/8 2 27/64 2 7/8 1 1 3/4 14

1 1/2 16 7/16 13 1/4 3 3/16 3 3/8 1 1/2 1 1/2 1 18

2 21 9/16 17 11/16 3 7/8 3 5/8 2 2 1 1/4 36

2 1/2 26 41/64 22 1/16 4 37/64 3 7/8 2 1/2 2 1/2 1 1/2 65

3 31 43/64 26 7/16 5 15/64 4 5/8 3 3 2 104

4 42 27/64 35 5/16 7 7/64 5 7/8 4 4 2 1/2 203

5 53 55/64 45 7/8 7 63/64 7 1/2 6 5 3 300

6 64 21/64 54 1/2 9 53/64 7 1/2 6 6 3 450

steam nozzle. Sizes 5 " and 6" still have the ductile iron body but tails

fabricated from steel. Details of construction and dimensions are shown

in Figures 6, and 7. The standard primary stage of a two-stage ejector

system (page 9), designated as Fig. 541, is constructed in the same

manner and externally follows the dimension in Table 1.

Sizes above 6" are made to special order and are generally 100%fabricated.

Ductile iron has strength characteristics similar to steel while retaining

many desirable features of cast iron. It is often used as a substitute for

steel. Units, however, can be supplied in steel, stainless steel and other

alloy utilizing barstock diffusers (see page 1).

S & K maintains sufficient parts inventory to assure component

availability in all standard sizes in ductile iron and stainless steel for fast

turnaround.

On special orders, ejectors can be supplied in Steel, Monel, Alloy 20,

Hastelloy, Titanium, Teflon, Haveg, Graphite (pages 5 and 6) and many

other materials.

Table 1. Sizes and Dimensions of Fig. 556 Ejectors (Standard Construction)

Note: Suction and discharge flanges are 150 Ib. ANSI.

D G

E

C

B A

F

Pressure

Connection

(Steam)

Suction

Connection

(air or other gases)

Discharge

Connection

(Mixture)

Pressure

Connection

(Steam)

Suction

Connection

(air or other gases)

Discharge

Connection

(Mixture)

Pipe Plug

Body

Removable

Nozzle

Removable

Venturi

Tail






FIG. 555H HAVEG CONSTRUCTION

Application

Designed for handling many solvents, as wellas acids and corrosive vapors, Fig. 555H

Steam Jet Ejectors are made from Haveg of

various types. Haveg resists rapid temperature

change and can be used continuously with

temperatures as high as 265° F. It is durable

and has excellent resistance to corrosion.

Construction

The standard unit is constructed of Haveg 61

with a graphite nozzle. Haveg 61 is a furfuryl

alcohol-formaldehyde resin with a non-

asbestos silicate filler and is used for body and

diffuser. A high grade of impervious Graphite is

used for the steam nozzle. Special applications

may require a different grade of Haveg

material.

The Fig. 555H Ejector has a one-piece molded

Haveg body and diffuser - eliminating a joint

between these parts, a steel steam chest and

a steam nozzle of Graphite. The bolts holding

the steam chest extend the full length of the

exhauster and fasten to the exhaust pipe. This

holds the body and diffuser in compression and

eliminates any tendency of the diffuser to break

away from the body.

Dimensions and sizes of 1” to 4” Fig. 555H

Haveg Ejectors are shown below. Haveg is a

plastic material which has been subjected to

thermal processing and pressure. Jet ejectors

made from this material in the grades available

are tough and durable and are resistant to

many acids, bases, and salts.

Fig. 9. Fig. 555H Haveg Ejector.

Fig. 10. Fig. 555H Steam Jet Ejector made o

Haveg. Nozzles are interchangeable with

those used in the Type 562 Ejecto

described on page 6.

Table 2. Sizes and Dimensions of Fig. 555H Ejector (Haveg Construction)

SizeNo.

(Inches)

Connections Dimensions Approx.Shipping

Wgt. (Lbs.)Suction DischargeSteam

InletE F G H

1 1 1/2 1 1/2 1/2 17 1/4 4 13 1/4 4 18

1 1/2 1 1/2 2 1/2 17 1/4 2 1/2 13 1/4 4 18

2 2 2 1/2 3/4 22 5/16 3 17 11/16 4 1/2 27

2 1/2 2 1/2 3 3/4 27 5/16 3 1/2 22 1/16 5 38

3 3 3 1 32 4 1/4 26 7/16 5 1/2 51

4 4 3 1 1/4 43 7/16 5 1/4 35 5/16 6 1/2 76

5 5 4 1 1/4

ON APPLICATION6 6 5 3

8 8 6 3

G

E

F

H

PressureConnection

(steam)

Suction

Connection

(air or

other gases)

Tie Bolts

Removable

Nozzle

Body and

Venturi

Tail Piece

(one piece

construction)

Discharge

Connection

(mixture)






FIG. 562/555G GRAPHITE CONSTRUCTION

Application

This type Steam Jet Ejector is designed to resist the corrosiveeffects of vapors from a large number of acid and salt solutions.

Construction

Specially constructed to make it non-porous and immune to the

effects of the vapors mentioned above, this Single-Stage

Ejector has a bronze steam chest, an impervious Graphite body,

nozzle, and tail bushing. External fiberglass armoring (Fig.

555G), which will add strength and assist in withstanding the

effects of corrosion, is provided in 4", 5 ", 6", and 8" sizes. (See

Fig. 13.)

Fig. 12. Fig. 562 Steam Jet Ejector, constructed

of impervious Graphite. Ejectors with 4" and

above suction and pressure connections havefiberglass armoring. Nozzles are

interchangeable with those used in the Fig.

555H Pump described on page 5.

Fig. 13. Sectional drawing showing design

and components of the fiberglass-armored

Fig. 555G Graphite Ejector (4” and above).

Fig. 14. Fig. 562 Ejectors with suction and

discharge connections of less than 4” are

metal armored as shown here.

Table 3. Sizes and Dimensions of Fig. 562/555G Ejectors (Graphite Construction)

Size

No.

(Inches)

Connections Dimensions Approx.

Shipping

Wgt. (Lbs.)Suction DischargeSteam

InletJ K L M

1 1/2 1 1/2 1 1/2 1/2 19 13/16 4 7/8 12 3/4 3 60

2 2 2 3/4 22 5/8 5 1/2 14 3/4 3 1/4 75

2 1/2 2 1/2 2 1/2 3/4 26 7/16 6 18 1/16 3 3/8 89

3 3 2 1/2 1 30 7/16 6 1/8 21 13/16 3 3/4 122

4 4 3 1 1/4 40 7/8 5 5/8 35 1/4 6 1/4 260

5 6 5 3 52 3/8 6 1/2 45 7/8 8 320

6 6 6 3 61 6 1/2 54 1/2 8 400

8 8 6 3 ON APPLICATION

M

K

L

J

Pressure

Connection

(Steam)

Body

Suction

Connection

(Air or Other Gases)

Venturi

Removable

Tail Piece

Discharge

Connection

(Mixture)

Pressure

Connection

(Steam)

Suction

Connection

(Air or Other Gases)

Body

Removable

Nozzle

Venturi

Removable

Tail Piece

Discharge

Connection

(Mixture)

A number of features make the design of this ejector

noteworthy. In addition, the Graphite is specially impregnated to

avoid leakage.

The steam chest is equipped with a stainless steel steam

strainer basket which is retained in place by a strainer plug. The

strainer plug is fitted with a pipe plug for easy inspection of

nozzle and strainer without removing steam lines or strainer

assembly. This plug may also be used to connect a steam

pressure gauge.

The diffuser and steam nozzle are accurately machined for

maximum steam economy. Dimensions and sizes from 1 1/2" to

6" are shown below.






FIG. 557/542 SINGLE-STAGE EJECTOR

RELIABLE VACUUM PERFORMANCE

Application

• Designed to cover a suction pressure range from 1” to

29” Hg absolute

• Unit re-designed to offer integral cast motive flange

• Standard components in stock to allow for fast

turnaround

• Units can be placed in series to attain high vacuum

levels

Construction

• Body: Investment cast in SST 316

• Tail: Investment cast or fab with choice of SST 316 or

carbon steel

• Motive connection available with 150# or 300# flanges

FIG. 557/542

5 3 5 1 6 4 2 7

MarkNo.

Description

1 Body

2 Tail

3 Nozzle

4 Capscrews

5 Pipe Plug

6 Gasket

7 Backing Flange (when needed)





UNIT

SIZE

UNIT DIMENSIONSE F

MOTIVE

SIZE

NET

WT

LBSA B C D G

1 14 1/2 8 7/8 5 5/8 2 7/8 1 1 3/4 23

1 1/2 20 13 1/4 6 3/4 3 3/8 1 1/2 1 1/2 1 27

2 25 1/4 17 11/16 7 9/16 3 5/8 2 2 1 1/2 54

2 1/2 30 1/4 22 1/16 8 3/16 3 7/8 2 1/2 2 1/2 1 1/2 83

3 36 3/4 26 7/16 10 5/16 4 5/8 3 3 2 126

4 47 5/16 35 5/16 12 5 7/8 4 4 2 1/2 222

5 59 45 7/8 12 7 1/2 6 5 3 343

6 69 1/2 54 1/2 12 7 1/2 6 6 3 493

Fig. 557/542 150# RF Motive Connection

Fig. 557/542 300# RF Motive Connection

D

E

C

DISCHARGE 150 # RF

SUCTION 150 # RF

MOTIVE 150 # RF

F

G

B

A

D

E

C

DISCHARGE 150 # RF

SUCTION 150 # RF

MOTIVE 300 # RF

F

G

B

A

UNIT

SIZE

UNIT DIMENSIONSE F

MOTIVE

SIZE

NET

WT

LBSA B C D G

1 14 1/2 8 7/8 5 5/8 2 7/8 1 1 3/4 23

1 1/2 20 13 1/4 6 3/4 3 3/8 1 1/2 1 1/2 1 27

2 25 1/4 17 11/16 7 9/16 3 5/8 2 2 1 1/2 54

2 1/2 30 1/4 22 1/16 8 3/16 3 7/8 2 1/2 2 1/2 1 1/2 83

3 36 3/4 26 7/16 10 5/16 4 5/8 3 3 2 126

4 47 5/16 35 5/16 12 5 7/8 4 4 2 1/2 222

5 57 7/8 45 7/8 12 7 1/2 6 5 3 343

6 66 1/2 54 1/2 12 7 1/2 6 6 3 493


FIG. 557/542 SINGLE-STAGE EJECTOR






Staging of ejectors becomes necessary for economical

operation as the required absolute suction pressure

decreases (see Fig. 3, page 3).

Based upon the use of auxiliary equipment, two and

three-stage ejectors can be either condensing or non-

condensing types. Four, five and six-stage units can also

be non-condensing, but usually are condensing types.

Condensing Type Ejectors (Fig. 16) have an

intercondenser between ejectors that reduces steam

consumption in later stages by (1) condensing first stage

operating steam and condensable vapors; and (2)

cooling the air and other non-condensables. The

intercondenser may be direct-contact or surface type,

arranged barometrically or low-level. Pages 10, 11 and 12

contain additional details on the Condensing Type

Ejector.

When the condenser is mounted at barometric elevation,

drainage is by gravity through a sealed tail leg so

condenser and suction lines will not flood if steam service

is interrupted or loss of vacuum occurs.

A ground-level arrangement suitable for many

applications is shown on page 14, Fig. 26. This type of

Fig. 16. Condensing Type.

Fig. 17. Non-Condensing Type.

steam jet ejector is ideal for use when service conditions

prohibit locating condensers at barometric height and

direct contact condensing is permitted.

Non-Condensing Type Ejectors (Fig. 17) have the first

stage ejector discharging directly into the suction of thesecond stage ejector and so on, using no condensers.

Compared to the Condensing Type Ejector, this

arrangement imposes a greater load on subsequent

stages, requiring more operating steam and larger units

following. Non-Condensing Type Ejectors are used where

condensers are not feasible, where initial cost is more

important than operating cost, or when service is to be

intermittent making operating cost a secondary

consideration.

Both Condensing Type Ejectors and Non-Condensing

Type Ejectors can be supplied with after-condensers. The

aftercondenser condenses the operating steam and anycondensable vapors before the non-condensables are

discharged to atmosphere.

Except for units of low capacity or those used for

intermittent service, condensing units are more

economical in operation than non-condensing types,

although initial cost may be higher. For photos of Multi-

Stage Non-Condensing Ejectors, refer to page 13.

Steam Inlet

Suction

H.V. Ejector

Tail Pipe

Wat er Inlet Steam Inlet

L.V.

Ejector

Tail Pipe

Water Inlet

Discharge

After

Cond.

Inter

Cond.

Steam Inlet

Suction

H.V. Ejector

Steam Inlet

L.V.

Ejector

Tail Pipe

Water Inlet

Discharge

After

Cond.






TWO-STAGE EJECTORS

Application

Two-Stage Steam Jet Ejectors have the same generalfield of application as the single stage units. They handle

both condensable and non-condensable gases or vapors,

as well as mixtures of the two. The general operating

range is between 5" Hg. abs. and 3 mm Hg. abs.

Depending on conditions, however, a single-stage unit

may be more economical at the top of the range and a

three-stage unit near the bottom.

Operation

In the two-stage assemblies, the suction mixture enters

the body of the primary stage, or High Vacuum (H.V.)

Ejector, Fig. 541, and is compressed from the required

suction to an intermediate pressure less than

atmospheric. The secondary stage or Low Vacuum (L.V.)

Ejector, Fig. 556 compresses from this point to

atmosphere, or to a point where it is desired to utilize the

ejector discharge.

Exact value of the intermediate pressure varies with the

operating conditions and the type of two-stage assembly.

The units have been designed for optimum inter-stage

pressure.

In condensing units, the inter-condenser functions as

previously described. This reduces the load on the low

vacuum ejector and reduces steam consumption. Theintercondenser may be a direct-contact barometric type,

a low level type, or surface type. These are discussed in

more detail on pages 12, 13 and 14.

In small size units, and where cooling water is not

economically available, the intercondenser may be

eliminated, resulting in a two-stage non-condensing unit.

When the suction load contains a large amount of

condensable vapors, it is sometimes possible to use a

surface or direct-contact pre-condenser, or pre-cooler to

reduce the load on the first stage ejector (Fig. 18). Also, if

it is objectionable to discharge the low vacuum exhauster

directly to atmosphere, an aftercondenser can be used to

condense the steam and other condensables, as well as

Iower the noise level. Direct-Contact Condensers for this

function are described in Bulletin 5AA.

Non-condensing two-stage units can be used when

conditions warrant this type of arrangement. A typical

arrangement is shown on page 13.

FIg. 18. Five Two-Stage Steam Jet Ejectors equipped with pre-condensers

are shown installed on the roof of a chemical plant. Tailpipes on

condensers are offset downward at a 45 degree angle to allow free flow of

discharge water.

Fig. 19. The Two-Stage Non-CondensingEjectors shown here are made of Haveg. The

piston-operated shut-off valve shown in the

suction line permits operation of the pump

from a central control panel.






THREE-STAGE EJECTORS

Application

Three-Stage Ejectors are recommended for applications where a two-stage unit will not provide low

enough suction pressure economically. Applicable

range is from 26 mm Hg. abs. to 0.8 mm Hg. abs. but

economics might dictate use of a Two-Stage Ejector at

the upper part of the range and a Four-Stage Ejector at

the lower end.

Operation

Three-Stage Condensing Steam Jet Ejectors consist of

a booster ejector, a booster condenser, and a Two-

Stage Ejector consisting of a high-vacuum ejector,

intercondenser, and low vacuum ejector. In some

applications another condenser (after-condenser) can

be used at the low vacuum ejector discharge.

The type condensers can be direct contact or surface

type arranged barometrically or low level. (See pages

10, 13 and 14).

The most economical type of three-stage ejector

system uses direct-contact, barometric, countercurrent

condensers which permit gravity drainage of the

condensate and condensing water and eliminate the

need for removal pumps. In cases where it is not

possible to install the unit at barometric height (about

34 feet), the low-level arrangement (Fig. 26, page 14)

can be used. In instances where contaminants are

introduced into the condensers and cannot be

discharged directly into drains, surface condensers areused to prevent discharge to drains and permit

recovery or treatment of the contaminants.

In condensing units, the booster ejector operates at

very high vacuum and discharges into a booster

condenser. Process and booster ejector steam is

condensed and the air and non-condensables are

cooled and pass over to the second stage ejector. This

continues through to the last stage (low vacuum

ejector) where they are compressed to atmosphere or,

if desired, into an aftercondenser. Cooling of non-

condensables reduces the load on succeeding

ejectors and minimizes steam consumption.

In general, units with direct-contact condensers require

less steam and cooling water than do those with

surface condensers.

Three-Stage Ejectors can also be of the non-

condensing type. They consist of a booster ejector,

high-vacuum ejector, and low-vacuum ejector, each

connected to the other by piping. From the third stage,

discharge is made to atmosphere or to a point where it

is desired to utilize the ejector discharge.

Fig. 20. Typical Three-

Stage Steam Jet

Ejector. This unit is

used in processing

vegetable oils.

Fig. 21. Three-Stage Steam Jet Ejector. The booster ejector diffuser should normall

be steam jacketed when design suction pressure is less than 4.6 mm Hg. abs.

B a r o m e t r i c

C o n d e n s e r

B a r o m e t r i c

C o n d e n s e r

High

Vacuum

Ejector

Low

Vacuum

Ejector

S t e a m

S e p a r a t o r

Booster

Ejector

Steam Water Water

Discharge

Discharge

Steam






FOUR, FIVE, AND SIX-STAGE EJECTORS

Application

Multi-Stage Ejectors have applications similar to thosedescribed on page 1 of this bulletin. These units are used

for applications where required suction pressures are

beyond the range of the ejectors previously described.

Generally, suction pressure ranges are as follows (note

overlap in bar chart. Fig. 3, page 3):

Four-Stage Ejectors—4 mm Hg. abs.

to 75 microns Hg. abs.

Five-Stage Ejectors—0.4 mm Hg. abs.


Six-Stage Ejectors—100 microns Hg. abs.


FOUR-STAGE EJECTORS

The four-stage unit consists of (1) a primary booster

ejector; (2) a secondary booster ejector; (3) a high

vacuum ejector; (4) a low vacuum ejector; and (5) usually

two condensers—one after the secondary booster ejector

and the other between the high vacuum and low vacuum

ejectors. The condenser between the high and low

vacuum ejectors is sometimes omitted, depending upon

application requirements. Direct contact or surface

condensers, arranged barometrically or at ground level,

can be used. The four-stage is similar to the three-stage

unit except that another booster ejector is added. In the

four-stage, the primary booster is steam-jacketed to

prevent build-up of ice on the diffuser internal bore.

In operation, the booster ejectors operate in series and

discharge into a booster condenser, which removes the

operating steam and condensable gases. From this point

operation is similar to the two-stage ejector.

Final selection and arrangement of four-stage units will

depend upon specific requirements.

FIVE AND SIX-STAGE EJECTORS

A typical Five-Stage Ejector is shown in Fig. 22. The five

and six-stage units are similar in appearance to the four-

stage ejector except that additional booster ejectors areadded. Suction pressure ranges are as indicated under

"application." The first two stages of these units are

usually steam-jacketed.

While four, five, and six-stage ejectors are usually

condensing types for reasons of efficiency and operating

economy, it is possible to employ non-condensing types.

Refer to S & K for information on operating characteristics

of such units.

Fig. 22. Five-Stage Steam Jet Ejector.

HighVacuum

Ejector

LowVacuum

Ejector

Tertiary

Booster

Primary

Booster

Secondary

Booster






EJECTORS WITH SURFACE CONDENSERS

Disposal of contaminated water is of growing concern in

process operations, particularly in the chemical industry. Where

an ejector system is drawing in contaminants, a condenser thatdischarges directly to the drain may not be used. In these

applications, ejectors using surface condensers are being

utilized more. The surface condenser prevents discharge to the

drain and permits recovery or treatment of undesirable wastes.

A steam jet system with surface condensers normally requires

more motive steam and condensing water than one with direct-

contact condensers. This is the most expensive type of multi-

NON-CONDENSING EJECTORS

Shown are several examples of multi-stage non-

condensing ejectors. This arrangement is generally

utilized in situations where a barometric leg or cooling

water is not readily available for an inter-condenser. They

also can be furnished in Haveg or Graphite construction

for corrosive applications. Non-condensing ejectors

provide the lowest initial capital equipment investment for

multi-stage systems.

Fig. 23. Two-Stage Steam

Jet Ejector System with

twin ejectors for each stage

and a surface type inter-and

aftercondenser.

Fig. 24. Two-Stage

Non-Condensing Ejector.

Fig. 25. Three-Stage

Non-Condensing Ejector.

Steam Inlet

Strainer

High Vacuum Ejectors

Isolating Valves

Inter-Condenser

After-Condenser

Isolating Valves

Condensing Water Outlet

Condensing Water Inlet

Low Vacuum Ejectors

Steam Valves

Isolating Valves

Air Inlet

Steam Valves

stage ejector. It can be mounted at barometric elevation, but

does not require this type of installation.

A typical multi-stage unit with twin ejectors for each stage and

surface type inter - and aftercondenser is shown in Fig. 23. The

purpose of the twin ejectors is to provide a spare set of ejectors

that can be brought into service in case repairs are necessary

on the other set. Isolating valves are used to allow removal of

an ejector without breaking vacuum.

In certain applications, the twin ejectors are selected to provide

more flexibility of operation under varying load conditions. In

this case, each ejector for each stage would be sized to handle

only half the load, so that the unit could operate at half-load with

only one ejector operating in each stage.






LOW-LEVEL EJECTOR SYSTEMS

Application

Low-Level Ejector Systems have applicationssimilar to those described on page 13. In cases

where it is not possible to install the condensing

portion at barometric height (34 feet), special

designs can be used for "direct-contact" (Fig.

26) and "shell and tube" (Fig. 27) type low-level

units. These units can be supplied as two-stage

through four-stage systems ready for operation

at job site simply by providing steam, water and

electrical connections.

Operation

DIRECT CONTACT LOW-LEVEL ARRANGEMENT

This type uses a direct contact condenser withan integral reservoir and a float-operated water

control valve to maintain a constant operating

head above the condensate removal pump.

Since heat is introduced by the process, it is

necessary to maintain proper condensing water

temperature by providing appropriate bleed and

make-up water.

SHELL AND TUBE LOW-LEVEL ARRANGEMENT

Standard shell and tube heat exchanger and a

pump operated water jet ejector are installed

below the exchanger to remove condensate.

The condensate removal system does not need

make-up cooling water after initial operation.

The steam jets supplied on both low-level types

are the same as supplied for barometric

installations.

Fig. 26. Four-Stage Low-Level Steam Jet

Ejector with an integral reservoir, a water

removal pump and level control.

Fig. 27. Two-Stage Low-Level arrangement

with shell and tube heat exchanger.

Fig. 28. With available head room at an

absolute minimum, S & K engineered this

three-stage low-level, condensing unit to

eliminate the need for a barometric leg.

Fig. 29. Standard Haveg Construction

with interconnecting tee and target

plate to take steam impingement.

CORROSION RESISTANT MULTI-STAGE

EJECTORS

Selection of suitable materials for the

specific pumping application is an

important consideration. To insure

minimum maintenance and replacement

costs, Multi-Stage Steam Jet Ejectors areavailable in many corrosion resistant

materials. Figures 29 and 30 show units

made of Haveg and Graphite. See page 2

for other special materials. Condensers are

frequently made of polyester fiberglass or

steel with neoprene lining.

Fig. 30. Standard Graphite Construction

(Haveg Intercondenser).






STEAM JET VACUUM BOOSTERS

Application

If large condensable vapor loads must be handled, such asthose from an evaporator or crystallizer, it is normally done with

a condenser followed by a single-stage or two-stage ejector.

The condenser condenses the vapor and the secondary unit

removes the saturated non-condensables and maintains the

vacuum.

The vacuum obtainable in a condenser is limited by the vapor

pressure of the injection water. If a higher vacuum is desired, a

Steam Jet Booster is provided to increase the vacuum to the

desired point. Boosters like this are used in multi-stage units.

The booster ejectors are large in proportion to the other ejectors

because of the magnitude of the vapor load they handle.

The function of the Steam Jet Vacuum Booster is to compress

the condensable and non-condensable vapors from the suctionvacuum to the intermediate vacuum maintained in the

condenser.

Fig. 31 shows a typical Steam Jet Vacuum Booster and its

construction. The vacuum booster is made of fabricated steel

Fig. 31. 30-inch vacuum

booster with carbon steel

body and stainless steel

diffuser.

Fig. 32. Typical

arrangement of a Three-Stage Steam Jet Ejector

with a Steam Jet Booster.

and has a nozzle which can be easily removed for examination

or cleaning without dismantling the booster body or pipe

connections. The nozzle can be cast or fabricated of special

materials if necessary.

The Fig. 533 Vacuum Booster is designed to handle large

quantities of condensable vapors plus relatively small quantities

of non-condensables in a pressure range of 5 to 25 mm Hg.

abs.

Fig. 32 shows a three-stage unit with a Steam Jet Booster

exhausting into a barometric condenser. Similar arrangements

are used extensively for vacuum distillation in oil refineries and

for other chemical processes, as well as for concentrating and

crystallizing liquids. Such an arrangement is also used to

remove vapors from a flash evaporator of a steam jet

refrigeration system.

Operation

The Jet Vacuum Booster is designed to operate with steam

pressures as low as 5 psig. In operation, the steam issues fromthe nozzle and creates a vacuum in the booster body. Suction

steam and vapors are drawn into the booster and entrained by

the operating pressure steam then discharged into the booster

condenser where steam and condensable vapors are

condensed.





APPLICATION CONSIDERATIONS

Operating Steam Pressure

All ejector nozzles are designed for a specific

steam flow and pressure. This pressure must be

maintained to insure stable and satisfactory

operation. Should the steam pressure drop below

the design pressure, the vacuum will drop and the

stability of performance will be upset. It is,

therefore, of the utmost importance when ordering

an ejector to specify the minimum steam pressure

available at any time at which the apparatus may

have to operate.

Should the steam pressure be increased above

the design pressure, the ejector will operate

satisfactorily with only a slight decrease in

capacity and with an increase in steam

consumption in direct proportion to the increase inthe absolute pressure. Where the operating steam

pressure is likely to vary over a wide range, we

recommend the installation of a suitable pressure

regulating valve in the steam line. Since moisture

in the steam will cause excess wear and erratic

operation, a steam separator is recommended.

Back Pressure

Standard ejectors are designed to operate against

a back pressure not exceeding 1 psig. It is

possible, however, to design them to operate

against higher back pressures, depending on thevacuum to be maintained and the available

operating steam pressure. However, ejectors

should not be operated against a back pressure

higher than that for which they are designed.

MEASUREMENT OF LOW ABSOLUTE PRESSURES

Following are precautions to use in connection with

steam jet ejectors:

1. Do not depend upon spring type vacuum gauges or

absolute pressure gauges involving one sealed leg

when perfect vacuum is assumed.

2. Gauge tubes should be clean and free of

contamination.

3. Gauge liquid should be clean and free of

contamination.

4. Barometer should be located near mercury

manometers.

5. If the operating point is 0.04 in. (1 mm) Hg. abs. or

less, rubber tubing should not be used. Copper

tubing or plastic tubing with a minimum bore of 3/8”should be used in this case.

The following gauges are recommended for use with the

vacuum pressures noted:

1. Absolute pressure of 4 in. Hg. to 0.5 in. Hg. Use a

mercury column or manometer using straight tube

with scale graduated to 0.1 in. and a vernier reading

to 0.01 in.

2. Absolute pressures of 12.7 mm Hg. to 1.0 mm Hg.

Use a Butyl Phthalate or similar differential oil

manometer.

3. Absolute pressures of less than 1.0 mm Hg. Use the

following:

a. A suitable McLeod gauge or oil manometer. The

McLeod gauge should be used without freezing trap

or mechanical dryer. The gauge and mercury must be

clean and the system tested to make certain there

are no leaks.

b. A suitable indicating vacuum gauge to be

connected in parallel with the McLeod gauge. This

indicating gauge can be a differential oil manometer

if the pressure is above 0.5 mm. Between 1 micron

and 1 mm, an ionization gauge or Piranni gauge maybe used. The purpose of the indicating gauge is to

show the observer when the pressure is steady

enough to take a reading with the more accurate

McLeod gauge.





DATA REQUIRED FOR QUOTATION

In order to select the type, size, and capacity of exhauster to meet

specific requirements, the following information should be supplied

with inquiries:

1. If multi-stage unit, specify type of unit desired (condensing

or non-condensing).

2. Fluid to be handled in Ib. per hour or standard cfm. If other

than air or water vapor, the molecular weight and specific

heat should be given. Vapor pressure of condensables

other than water vapor is also required.

3. Materials of construction required. If this is in doubt, an

analysis of the suction fluid should be presented to aid in

making the proper selection.

4. Temperature of suction fluid at exhauster inlet.

5. Pressure desired at ejector suction, in inches, millimeters,or microns of mercury absolute.

6. Minimum pressure of operating steam stating whether

steam is dry, saturated or superheated, giving degree of

superheat, if any.

7. Maximum temperature of water available and minimum

pressure of condenser inlet.

8. State whether final stage ejector is to operate against

atmospheric pressure or a higher back pressure, and if

so, what pressure.

9. Normal barometer reading at installation.

10. Type of condenser desired—direct contact or surface type

(if required).

11. Type of installation desired-barometric or low level. If low

level, state electrical code the removal pump must meet.





Fig. 33. These two steam jet ejectors serve as part of the pressure recovery system at the U.S. Army'shigh energy laser system test facility in White Sands, New Mexico. They are each 97 feet long with 96

inch diameter end-suction connections, and are among the largest ever manufactured anywhere.

Each ejector handles a large quantity of low molecular weight gas at 120 Torr using the equivalent of

1.044 million pounds of steam per hour at 150 psig during the 14-second cycle.

Fig. 34. This compact, twin-element, two-stage steam jet ejector saves space in a nuclear power

station. Each first-stage ejector discharges into a separate intercondenser, while both second-

stage ejectors discharge into a common after condenser. Twin element designs of the type

shown provide uninterrupted service. Either element can be taken out of service for periodic

inspection and cleaning while the other continues to function.

ISO

9001:2000

Certified

022707



CATALOGO GENERAL

DE

EYECTORES



EL EYECTOR

El eyector es una bomba estática, sin partes mecánicas en movimiento, caracterizado

por:

- seguridad de funcionamiento

- fiabilidad de funcionamiento con el tiempo

- amplia gama de ejecuciones

- libre de mantenimiento

- sin partes eléctricas

- fácil instalación incluso en sitios de difícil acceso

El eyector es una bomba estática, constituido principalmente por una tobera y un

difusor de sección cónica.

Entrada

Fluido motor

Salida mezcla

Entrada fluido aspirado

El eyector utiliza la energía del fluido primario en presión, llamado también fluido

motor, que puede ser agua, aire, vapor u otro tipo de fluido, para aspirar, mezclar y

comprimir el fluido secundario, también llamado fluido aspirado, por ejemplo agua,

aire, vapor u otros fluidos, según el principio de Bernoulli.

Ecuación de Bernoulli:

v p

z co n st 2

2

+ + = ρ



EYECTORES CON FLUIDO MOTOR AGUAu otro líquido

Para elevación de líquidos

Para mezclas de líquidos

Para disolución de ácidos o bases

Para procesos de vacío en continuo

Para cebado de bombas y sifones

Para la compresión de gases

Para el transporte de sólidos

Para la aspiración delíquidos

Para hacer vacío Para la compresión de gas Para el transporte desólidos

Datos a facilitar para la elaboración de una oferta:

Características del fluido aspirado

- tipo de fluido, peso molecular, peso específico, etc.

- caudal a aspirar

- presión de aspiración- temperatura de aspiración

- presión de descarga o altura de elevación

Características del fluido motor


- presión motriz disponible

- temperatura del fluido motor

En el caso de eyectores de arranque:

- volumen a evacuar

- presión inicial

- presión final

- tiempo de arranque

Para fluidos corrosivos



EYECTORES CON FLUIDO MOTOR AIREu otro gas

Para la realización de vacío en continuo

Para la realización de vacío de cebado

Para ventilación

Para la compresión de gas

Para aumentar el grado de vacío de una bomba de anillo líquido

Para vacío Fluido motor aire atmosférico En un sistema de vacío con bomba



- tipo de fluido, peso molecular, peso específico etc.

- caudal a aspirar

- presión de aspiración

- temperatura de aspiración

- presión de descarga


- tipo de fluido, peso molecular, peso específico, etc.- presión motriz disponible



- volumen a evacuar

- presión inicial

- presión final

- tiempo de arranquePara cebado mediante vacío



EYECTORES CON CHORRO DE VAPOR DE AGUAu otro vapor

Para la realización de vacío en continuo

Para la realización de cebado mediante vacío

Para ventilación

Para la compresión de gas

Para la elevación de líquidos

Para vacío Para un rápido cebado Para vacío Con difusorcalefaccionado




- caudal a aspirar

- presión de aspiración- temperatura de aspiración

- presión de descarga o altura de elevación






- volumen a evacuar

- presión inicial

- presión final

- tiempo de arranquePara la elevación de líquidos



CALENTADORESde líquido con mezcla directa de vapor

Para instalaciones en línea

Para instalaciones fuera del depósito

Para instalaciones en el interior del depósito


Características del fluido a calentar


- caudal a calentar

- temperatura a la entrada

- temperatura a la salida

- presión a la entrada

-presión a la salidaCaracterísticas del fluido calefactor


- presión del fluido calefactor

- temperatura del fluido calefactor

En el caso de calentamiento de tanques:

- volumen a calentar

- temperatura inicial

- temperatura final

- tiempo de calentamiento

DESOBRECALENTADORESde vapor con mezcla directa de líquido

Para instalaciones en líneaDatos a facilitar para la elaboración de una oferta:

Características del fluido a desobrecalentar


- caudal a calentar

- temperatura a la entrada

- temperatura a la salida

- presión a la entrada

Características del fluido de refrigeración


- temperatura disponible a la entrada

- presión disponible a la entrada



GRUPOS DE VACÍOcon eyectores de vapor de varias etapas

Totalmente con eyectores

Sistema mixto con bomba final de anillo líquido

Con condensadores de mezcla

Con condensadores de superficies

De cuatro etapas con termocompresore intercambiadores de mezcla

De dos etapas con bombas de vacio eintercambiador de superficie




- caudal a aspirar

- presión de aspiración

- temperatura de aspiración

- presión de descarga

Características del fluido motor- tipo de fluido, peso molecular, peso específico, etc.



Características del fluido de refrigeración


- presión de entrada

- temperatura de entrada


- volumen a evacuar

- presión inicial

- presión final

- tiempo de arranque De dos etapas con condensador

de mezcla

De tres etapas con

condensadores de mezcla





H . El -Dessouky et al . / Chemical Engineering and Processing 41 (2002) 551– 561552

Ejectors have very low thermal ef ficiency.

Applications of jet ejectors include refrigeration, air

conditioning, removal of non-condensable gases, trans-

port of solids and gas recovery. The function of the jet

ejector differs considerably in these processes. For ex-

ample, in refrigeration and air conditioning cycles, the

ejector compresses the entrained vapor to higher pres-

sure, which allows for condensation at a higher temper-

ature. Also, the ejector entrainment process sustains the

low pressure on the evaporator side, which allows

evaporation at low temperature. As a result, the cold

evaporator fluid can be used for refrigeration and cool-

ing functions. As for the removal of non-condensable

gases in heat transfer units, the ejector entrainment

process prevents their accumulation within condensers

or evaporators. The presence of non-condensable gases

in heat exchange units reduces the heat transfer ef fi-

ciency and increases the condensation temperature be-

cause of their low thermal conductivity. Also, the

presence of these gases enhances corrosion reactions.

However, the ejector cycle for cooling and refrigerationhas lower ef ficiency than the MVC units, but their

merits are manifested upon the use of low grade energy

that has limited effect on the environment and lower

cooling and heating unit cost.

Although the construction and operation principles

of jet ejectors are well known, the following sections

provide a brief summary of the major features of

ejectors. This is necessary in order to follow the discus-

sion and analysis that follow. The conventional steam

jet ejector has three main parts: (1) the nozzle; (2) the

suction chamber; and (3) the diffuser (Fig. 1). The

nozzle and the diffuser have the geometry of converg-

ing/diverging venturi. The diameters and lengths of

various parts forming the nozzle, the diffuser and the

suction chamber, together with the stream flow rate and

properties, define the ejector capacity and performance.

The ejector capacity is defined in terms of the flow rates

of the motive steam and the entrained vapor. The sum

of the motive and entrained vapor mass flow rates gives

the mass flow rate of the compressed vapor. As for the

ejector performance, it is defined in terms of entrain-

ment, expansion and compression ratios. The entrain-ment ratio (w) is the flow rate of the entrained vapor

Fig. 1. Variation in stream pressure and velocity as a function of location along the ejector.



H . El -Dessouky et al . / Chemical Engineering and Processing 41 (2002) 551– 561 553

divided by the flow rate of the motive steam. As for the

expansion ratio (Er), it is defined as the ratio of the

motive steam pressure to the entrained vapor pressure.

The compression ratio (Cr) gives the pressure ratio of

the compressed vapor to the entrained vapor.

Variations in the stream velocity and pressure as a

function of location inside the ejector, which are shown

in Fig. 1, are explained below:

The motive steam enters the ejector at point ( p) with

a subsonic velocity.

As the stream flows in the converging part of the

ejector, its pressure is reduced and its velocity in-

creases. The stream reaches sonic velocity at the

nozzle throat, where its Mach number is equal to one.

The increase in the cross section area in the diverging

part of the nozzle results in a decrease of the shock

wave pressure and an increase in its velocity to

supersonic conditions.

At the nozzle outlet plane, point (2), the motive steam

pressure becomes lower than the entrained vapor

pressure and its velocity ranges between 900 and 1200m/s.

The entrained vapor at point (e) enters the ejector,

where its velocity increases and its pressure decreases

to that of point (3).

The motive steam and entrained vapor streams may

mix within the suction chamber and the converging

section of the diffuser or it may flow as two separate

streams as it enters the constant cross section area of

the diffuser, where mixing occurs.

In either case, the mixture goes through a shock

inside the constant cross section area of the diffuser.

The shock is associated with an increase in themixture pressure and reduction of the mixture veloc-

ity to subsonic conditions, point (4). The shock

occurs because of the back pressure resistance of the

condenser.

As the subsonic mixture emerges from the constant

cross section area of the diffuser, further pressure

increase occurs in the diverging section of the dif-

fuser, where part of the kinetic energy of the mixture

is converted into pressure. The pressure of the emerg-

ing fluid is slightly higher than the condenser pres-

sure, point (c).

Summary for a number of literature studies on ejectordesign and performance evaluation is shown in Table 1.

The following outlines the main findings of these studies:

Optimum ejector operation occurs at the critical

condition. The condenser pressure controls the loca-

tion of the shock wave, where an increase in the

condenser pressure above the critical point results in

a rapid decline of the ejector entrainment ratio, since

the shock wave moves towards the nozzle exit. Oper-

ating at pressures below the critical points has negli-

gible effect on the ejector entrainment ratio.

At the critical condition, the ejector entrainment ratio

increases at lower pressure for the boiler and con-

denser. Also, higher temperature for the evaporator

increases the entrainment ratio.

Use of a variable position nozzle can maintain the

optimum conditions for ejector operation. As a re-

sult, the ejector can be maintained at critical condi-

tions even if the operating conditions are varied.

Multi-ejector system increases the operating rangeand improves the overall system ef ficiency.

Ejector modeling is essential for better understanding

of the compression process, system design and perfor-

mance evaluation. Models include empirical correla-

tions, such as those by Ludwig [1], Power [2] and

El-Dessouky and Ettouney [3]. Such models are lim-

ited to the range over which it was developed, which

limits their use in investigating the performance of

new ejector fluids, designs or operating conditions.

Semi-empirical models give more flexibility in ejector

design and performance evaluation [4,5]. Other ejec-

tor models are based on fundamental balance equa-tions [6].

This study is motivated by the need for a simple

empirical model that can be used to design and evaluate

the performance of steam jet ejectors. The model

is based on a large database extracted from several

ejector manufacturers and a number of experimental

literature studies. As will be discussed later, the model

is simple to use and it eliminates the need for iterative

procedures.

2. Mathematical model

The review by Sun and Eames [7] outlined the devel-

opments in mathematical modeling and design of jet

ejectors. The review shows that there are two basic

approaches for ejector analysis. These include mixing of

the motive steam and entrained vapor, either at constant

pressure or at constant area. Design models of stream

mixing at constant pressure are more common in litera-

ture because the performance of the ejectors designed by

this method is more superior to the constant area

method and it compares favorably against experimental

data. The basis for modeling the constant pressure

design procedure was initially developed by Keenan [6].Subsequently, several investigators have used the model

for design and performance evaluation of various types

of jet ejectors. This involved a number of modifications

in the model, especially losses within the ejector and

mixing of the primary and secondary streams. In this

section, the constant pressure ejector model is devel-

oped. The developed model is based on a number of

literature studies [8 – 11].

The constant pressure model is based on the following

assumptions:




Table 1

Summary of literature studies on ejector design and performance

Boiler, evaporator and condenserFluidReference Conclusion

temperature (°C)

60 – 100; 5 – 18; 40 – 50[19] Basis for refrigerant selection for solar system, system performanceR-113

increased with increasing boiler and evaporator temperatures and

decreasing condenser temperature.

R-113; R-114;[20] 80 – 95; 5 – 13; 25 – 45 Comparison of ejector and refrigerant performance. Dry, wet andisentropic fluids. Wet fluid damage ejectors due phase change duringR-142b; R-718

isentropic expansion. R-113 (dry) has the best performance and

R142b (wet) has the poorest performance.

86; −8; 30[21,22] Increase in ejector performance using mechanical compressionR-114

booster.

120 – 140; 5 – 10; 30 – 65Water Choking of the entrained fluid in the mixing chamber affects system[8]

performance. Maximum COP is obtained at the critical flow

condition.

120 – 140; 5 – 10; 30 – 60[13] Effect of varying the nozzle position to meet operating condition.Water

Increase in COP and cooling capacity by 100%.

70 – 100; 6 – 25; 42 – 50[23] Entrainment ratio is highly affected by the condenser temperatureR-113

especially at low evaporator temperature.

82.2 – 182.2; 10; 43.3 Entrainment ratio is proportional to boiler temperature.R-11[24]

R-114 90; 4; 30 Combined solar generator and ejector air conditioner. More ef ficient[25,26]

system requires multi-ejector and cold energy storage (cold storage in

either phase changing materials, cold water or ice).

[27] −15; 30 Modeling the effect of motive nozzle on system performance, inR-134A

which the ejector is used to recover part of the work that would be

lost in the expansion valve using high-pressure motive liquid.

[28] 100 – 165; 10; 30 – 45 Combined solar collector, refrigeration and seawater desalinationWater

system. Performance depends on steam pressure, cooling water

temperature and suction pressure.

Water[4] Developed a new ejector theory in which the entrained fluid is

choked, the plant scale results agree with this theory. Steam jet

refrigeration should be designed for the most often prevailing

conditions rather than the most severe to achieve greater overall

ef ficiency.

Water[29] – Model of multistage steam ejector refrigeration system using annular

ejector in which the primary fluid enters the second stage at annular

nozzle on the sidewall. This will increase static pressure for

low-pressure stream and mixture and reduce the velocity of the

motive stream and reduce jet mixing losses shock wave formation

losses.

R11; R113;[24] 93.3; 10; 43.3 Measure and calculate ejector entrainment ratio as a function of

boiler, condenser and evaporator temperatures. Entrainment ratioR114

decreases for off design operation and increases for the two stage

ejectors.

[30] R113; R114; 120 – 140; 65 – 80 Effect of throat area, location of main nozzle and length of the

R142b constant area section on backpressure, entrainment ratio andcompression ratio.

Mathematical model use empirical parameters that depend solely on[5]

geometry. The parameters are obtained experimentally for various

types of ejectors.

5; −12, −18; 40[31] Combined ejector and mechanical compressor for operation of R134a

domestic refrigerator-freezer increases entrainment ratio from 7 to

12.4%. The optimum throat diameter depends on the freezer

temperature

80; 5; 30[9] Performance of HR-123 is similar to R-11 in ejector refrigeration.R11; HR-123

Optimum performance is achieved by the use of variable geometry

ejector when operation conditions change.




1. The motive steam expands isentropically in the

nozzle. Also, the mixture of the motive steam and

the entrained vapor compresses isentropically in the

diffuser.

2. The motive steam and the entrained vapor are

saturated and their velocities are negligible.

3. Velocity of the compressed mixture leaving the ejec-

tor is insignificant.

4. Constant isentropic expansion exponent and theideal gas behavior.

5. The mixing of motive steam and the entrained

vapor takes place in the suction chamber.

6. The flow is adiabatic.

7. Friction losses are defined in terms of the isentropic

ef ficiencies in the nozzle, diffuser and mixing

chamber.

8. The motive steam and the entrained vapor have the

same molecular weight and specific heat ratio.

9. The ejector flow is one-dimensional and at steady

state conditions.

The model equations include the following: Overall material balance

mp+me=mc (1)

where m is the mass flow rate and the subscripts c, e

and p, define the compressed vapor mixture, the

entrained vapor and the motive steam or primary

stream.

Entrainment ratio

w=me/mp (2)

Compression ratio

Cr=Pc/Pe (3)

Expansion ratio

Er=Pp/Pe (4)

Isentropic expansion of the primary fluid in the

nozzle is expressed in terms of the Mach number of

the primary fluid at the nozzle outlet plane

M p2= 2n

−1 Pp

P2(−1/)

−1n (5)

where M is the Mach number, P is the pressure and

is the isentropic expansion coef ficient. In the above

equation, n is the nozzle ef ficiency and is defined as

the ratio between the actual enthalpy change and the

enthalpy change undergone during an isentropic

process.

Isentropic expansion of the entrained fluid in the

suction chamber is expressed in terms of the Mach

number of the entrained fluid at the nozzle exit plane

M e2= 2

−1

Pe

P2

(−1/)

−1n

(6)

The mixing process is modeled by one-dimensional

continuity, momentum and energy equations. These

equations are combined to define the critical Mach

number of the mixture at point 5 in terms of the

critical Mach number for the primary and entrainedfluids at point 2

M 4*= M p2* +wM e2* T e/T p

(1+w)(1+wT e/T p)(7)

where w is the entrainment ratio and M * is the ratio

between the local fluid velocity to the velocity of

sound at critical conditions.

The relationship between M and M * at any point in

the ejector is given by this equation

M*= M 2(+1)

M 2(−1)+2 (8)

Eq. (8) is used to calculate M e2* , M p2* , M 4 Mach number of the mixed flow after the shock

wave

M 5=

M 42+

2

(−1)

2

(−1) M 4

2−1

(9)

Pressure increase across the shock wave at point 4

P5

P4

=1+M 4

2

1+M 52

(10)

In Eq. (10) the constant pressure assumption implies

that the pressure between points 2 and 4 remains

constant. Therefore, the following equality con-

straint applies P2=P3=P4.

Pressure lift in the diffuser

Pc

P5

=d(−1)

2 M 5

2+1n(/−1)

(11)

where d is the diffuser ef ficiency.

The area of the nozzle throat

A1=mp

Pp

RT p

n

+1

2

(+1)/(−1)

(12)

The area ratio of the nozzle throat and diffuser

constant area

A1

A3

=Pc

Pp

1

(1+w)(1+w(T e/T p))

1/2P2

Pc

1/1−P2

Pc

(−1)/1/2 2

+1

1/(−1)1−

2

+1

1/2 (13)




The area ratio of the nozzle throat and the nozzle

outlet

A2

A1

= 1

M p2

2

2

(+1

1+

(−1)

2 M p

2

2 (+1)/(−1)

(14)

3. Solution procedure

Two solution procedures for the above model are

shown in Fig. 2. Either procedure requires iterative

calculations. The first procedure is used for system

design, where the system pressures and the entrainment

ratio is defined. Iterations are made to determine the

pressure of the motive steam at the nozzle outlet (P2) that

gives the same back pressure (Pc). The iteration sequence

for this procedure is shown in Fig. 2(a) and it includes

the following steps:

Define the design parameters, which include the en-

trainment ratio (w), the flow rate of the compressed

vapor (mc) and the pressures of the entrained vapor,

compressed vapor and motive steam (Pe, Pp, Pc).

Define the ef ficiencies of the nozzle and diffuser (n,

d).

Calculate the saturation temperatures for the com-

pressed vapor, entrained vapor and motive steam,

which include T c, T p, T e, using the saturation temper-

ature correlation given in the appendix.

As for the universal gas constant and the specific heatratio for steam, their values are taken as 0.462 and 1.3.

The flow rates of the entrained vapor (me) and motive

steam (mp) are calculated from Eqs. (1) and (2).

A value for the pressure at point 2 (P2) is estimated

and Eqs. (5) – (11) are solved sequentially to obtain the

pressure of the compressed vapor (Pc).

The calculated pressure of the compressed vapor is

compared to the design value.

A new value for P2 is estimated and the previous step

is repeated until the desired value for the pressure of

the compressed vapor is reached.

Fig. 2. Solution algorithms of the mathematical model. (a) Design procedure to calculate area ratios. (b) Performance evaluation to calculate w .




The ejector cross section areas (A1, A2, A3) and the

area ratios (A1/A3 and A2/A1) are calculated from

Eqs. (12) – (14).

The second solution procedure is used for perfor-

mance evaluation, where the cross section areas and the

entrainment and motive steam pressures are defined.

Iterations are made to determine the entrainment ratio

that defines the ejector capacity. The iteration sequence

for this procedure is shown in Fig. 2(b) and it includesthe following steps:

Define the performance parameters, which include

the cross section areas (A1, A2, A3), the pressures of

the entrained vapor (Pe) and the pressure of the

primary stream (Pp).

Define the ef ficiencies of the nozzle and diffuser (n,

d).

Calculate the saturation temperatures of the primary

and entrained streams, T p and T e, using the satura-

tion temperature correlation given in the appendix.

As for the universal gas constant and the specific

heat ratio for steam, their values are taken as 0.462and 1.3.

Calculate the flow rate of the motive steam and the

properties at the nozzle outlet, which include mp, P2,

M e2, M p2. These are obtained by solving Eqs. (5),

(6), (12) and (14).

An estimate is made for the entrainment ratio, w.

This value is used to calculate other system parame-

ters defined in Eqs. (7) – (11), which includes M e2

* ,

M p2

* , M 4*, M 4, M 5, P5, Pc.

A new estimate for w is obtained from Eq. (13).

The error in w is determined and a new iteration is

made if necessary.

The flow rates of the compressed and entrained

vapor are calculated from Eqs. (1) and (2).

4. Semi-empirical model

Development of the semi-empirical model is thought

to provide a simple method for designing or rating of

steam jet ejectors. As shown above, solution of the

mathematical model requires an iterative procedure.

Also, it is necessary to define values of n and d. The

values of these ef ficiencies widely differ from one study

to another, as shown in Table 2. The semi-empiricalmodel for the steam jet ejector is developed over a wide

range of operating conditions. This is achieved by using

three sets of design data acquired from major ejector

manufacturers, which includes Croll Reynolds, Graham

and Schutte – Koerting. Also, several sets of experimen-

tal data are extracted from the literature and are used

in the development of the empirical model. The semi-

empirical model includes a number of correlations to

calculate the entrainment ratio (w), the pressure at the

nozzle outlet (P2) and the area ratios in the ejector

Table 2

Examples of ejector ef ficiencies used in literature studies

nReference md

0.9[27] 0.75

[32] 0.8 0.8

[33] 0.85 0.85

0.7 – 10.7 – 1[31]

[10] 0.8 – 1 0.8 – 1

0.85 – 0.98[24] 0.65 – 0.85

0.950.850.85[8]

0.75[34] 0.9

(A2/A1) and (A1/A3). The correlation for the entrain-

ment ratio is developed as a function of the expansion

ratio and the pressures of the motive steam, the en-

trained vapor and the compressed vapor. The correla-

tion for the pressure at the nozzle outlet is developed as

a function of the evaporator and condenser pressures.

The correlations for the ejector area ratios are defined

in terms of the system pressures and the entrainment

ratio. Table 3 shows a summary of the ranges of the

experimental and the design data. The table also in-

cludes the ranges for the data reported by Power [12].

A summary of the experimental data, which is used

to develop the semi-empirical model is shown in Table

4. The data includes measurements by the following

investigators:

Eames et al. [8] obtained the data for a compression

ratio of 3 – 6, expansion ratio 160 – 415 and entrain-

ment ratio of 0.17 – 0.58. The measurements are ob-

tained for an area ratio of 90 for the diffuser and the

nozzle throat.

Munday and Bagster [4] obtained the data for acompression ratio of 1.8 – 2, expansion ratio of 356 –

522 and entrainment ratio of 0.57 – 0.905. The mea-

surements are obtained for an area ratio of 200 for

the diffuser and the nozzle throat.

Aphornratana and Eames [13] obtained the data for

a compression ratio of 4.6 – 5.3, expansion ratio of

309.4 and entrainment ratio of 0.11 – 0.22. The mea-

surements are obtained for an area ratio of 81 for

the diffuser and the nozzle throat.

Bagster and Bresnahan [14] obtained the data for a

compression ratio of 2.4 – 3.4, expansion ratio of

165 – 426 and entrainment ratio of 0.268 – 0.42. Themeasurements are obtained for an area ratio of 145

for the diffuser and the nozzle throat.

Sun [15] obtained the data for a compression ratio of

2.06 – 3.86, expansion ratio of 116 – 220 and entrain-

ment ratio of 0.28 – 0.59. The measurements are ob-

tained for an area ratio of 81 for the diffuser and the

nozzle throat.

Chen and Sun [16] obtained the data for a compres-

sion ratio of 1.77 – 2.76, expansion ratio of 1.7 – 2.9

and entrainment ratio of 0.37 – 0.62. The measure-




ments are obtained for an area ratio of 79.21 for the

diffuser and the nozzle throat.

Arnold et al. [17] obtained the data for a compres-

sion ratio of 2.47 – 3.86, expansion ratio of 29.7 –

46.5, and entrainment ratio of 0.27 – 0.5.

Everitt and Riffat [18] obtained the data for a com-

pression ratio of 1.37 – 2.3, expansion ratio of 22.6 –

56.9 and entrainment ratio of 0.57.

The correlation for the entrainment ratio of chokedflow or compression ratios above 1.8 is given by

W=aErbPecP c

d (e+ fPp

g)

(h+ iPc j )

(15)

Similarly, the correlation for the entrainment ratio of

un-choked flow with compression ratios below 1.8 is

given by

W=aErbPecP c

d (e+ f ln(Pp))

( g +h ln(Pc)) (16)

The constants in Eqs. (15) and (16) are given asfollows

Entrainment ratio Entrainment ratio

correlation choked correlation non-choked

flow (Eq. (15); Fig. 3) flow (Eq. (16), Fig. 4)

a 0.65 −1.89×10−5

−1.54b −5.32

c 1.72 5.04

9.05×10−26.79v10−2d

22.82e 22.09

f 4.21×10−4 −6.130.82 g 1.34

h −3.37×10−59.32

1.28×10−1 − j

− j 1.14

R2 0.85 0.79

Fitting results against the design and experimental

data are shown in Figs. 3 and 4, respectively. The

results shown in Fig. 3 cover the most commonly used

range for steam jet ejectors, especially in vacuum and

vapor compression applications. As shown in Fig. 3,

the fitting result is very satisfactory for entrainment

ratios between 0.2 and 1. This is because the major part

of the data is found between entrainment ratios clus-

tered over a range of 0.2 – 0.8. Examining the experi-

mental data fit shows that the major part of the data fit

is well within the correlation predictions, except for a

small number of points, where the predictions have

large deviations.The correlations for the motive steam pressure at the

nozzle outlet and the area ratios are obtained semi-em-

pirically. In this regard, the design and experimental

data for the entrainment ratio and system pressures are

used to solve the mathematical model and to calculate

the area ratios and motive steam pressure at the nozzle

outlet. The results are obtained for ef ficiencies of 100%

for the diffuser, nozzle and mixing and a value of 1.3

for . The results are then correlated as a function of

the system variables. The following relations give the

correlations for the choked flow:

P2=0.13 P e0.33Pc

0.73 (17)

A1/A3=0.34 P c1.09Pp

−1.12w−0.16 (18)

A2/A1=1.04 P c−0.83Pp

0.86w−0.12 (19)

The R 2 for each of the above correlations is above 0.99.

Similarly, the following relations give the correlations

for the un-choked flow:

P2=1.02 P e−0.000762P c

0.99 (20)

A1/A3=0.32 P c1.11Pp

−1.13w−0.36 (21)

A2/A1=1.22 P c

−0.81

Pp

0.81

w

−0.0739

(22)The R2 values for the above three correlations are

above 0.99.

The semi-empirical ejector design procedure involves

sequential solution of Eqs. (1) – (14) together with Eq.

(17) or Eq. (20) (depending on the flow type, choked or

non-choked). This procedure is not iterative in contrast

with the procedure given for the mathematical model in

the previous section. As for the semi-empirical perfor-

mance evaluation model, it involves non-iterative solu-

tion of Eqs. (1) – (14) together with Eq. (15) or Eq. (16)

for choked or non-choked flow, respectively. It should

be stressed that both solution procedures are indepen-

Table 3

Range of design and experimental data used in model development

ErSource Cr Pe (kPa) Pc (kPa) Pp (kPa) w

1.6 – 526.1 0.872 – 121.3Experimental 2.3 – 224.11.4 – 6.19 38.6 – 1720 0.11 – 1.132

1.008 – 3.73 0.1 – 484.09 – 2132.27790.8 – 2859.2266.85 – 2100.81.36 – 32.45Schutte – Koerting

1.25 – 4.24 4.3 – 429.4 3.447 – 124.1Croll – Rynolds 446.06 – 1480.27 6.2 – 248.2 0.1818 – 2.5

1.174 – 4.04 4.644 – 53.7 27.58 – 170.27 790.8 – 1480.27 34.47 – 301.27Graham 0.18 – 3.23

1.047 – 5.018 2 – 1000 2.76 – 172.37 3.72 – 510.2 344.74 – 2757.9Power 0.2 – 4




Table 4

Summary of literature experimental data for steam jet ejectors

Pe (kPa) Pc (kPa) Pp/Pe Pc/PePp (kPa) wAd/At Reference

1.2390 3.8198.7 161.8 3.09 0.59 [8]

1.23 4.2 189.1232.3 3.42 0.54 [8]

270.3 1.23 4.7 220.1 3.83 0.47 [8]

1.23 5.3 255.1 4.31 0.39 [8]313.31.23 6 294.4 4.89361.6 0.31 [8]

1.04 3.6 191.690 3.47198.7 0.5 [8]

1.04 4.1 223.9232.3 3.95 0.42 [8]

270.3 1.04 4.6 260.7 4.44 0.36 [8]

1.04 5.1 302.1313.3 4.91 0.29 [8]

361.6 1.04 5.7 348.7 5.49 0.23 [8]

0.87 3.4 227.7 3.89 0.490 [8]198.7

0.87 3.7 266.2232.3 4.24 0.34 [8]

0.87 4.4 309.8 5.04270.3 0.28 [8]

0.87 5.1 359313.3 5.85 0.25 [8]

361.6 0.87 5.4 414.4 6.19 0.18 [8]

1.59 3.2 521.7200 2.0834 0.58 [4]

400 1.59 3.07 250.2 1.92 1.13 [4]1.71 3.67 392.3 2.15669 0.58 [4]

1.59 3.51 526.1841 2.19 0.51 [4]

1.94 3.38 356 1.74 0.86690 [4]

1.94 3.51 356 1.81690 0.91 [4]

81 270 0.87 4.1 309.5 4.7 0.22 [13]

0.87 4.2 309.5270 4.8 0.19 [13]

270 0.87 4.4 309.5 5.04 0.16 [13]

0.87 4.5 309.5 5.16 0.14270 [13]

0.87 4.7 309.5 5.39270 0.11 [13]

1.55 5.3 426.5145 3.42660 0.27 [14]

1.55 5.3 373.5578 3.42 0.31 [14]

516 1.58 5.3 326.9 3.36 0.35 [14]

1.57 5.03 280.6440 3.21 0.38 [14]

381 1.59 4.77 239.9 3 0.42 [14]1.62 4.23 192.6 2.61 0.46312 [14]

1.68 4.1 165.1 2.44278 0.42 [14]

1.23 2.53 116.881 2.06143.4 0.59 [15]

1.23 2.67 137.8169.2 2.17 0.51 [15]

198.7 1.23 3.15 161.8 2.56 0.43 [15]

1.23 4 189.1232.3 3.26 0.35 [15]

270.3 1.23 4.75 220.1 3.87 0.29 [15]

57.7 1431720 29.7 2.47 0.5 [17]

51.4 143 33.51720 2.78 0.4 [17]

45.5 143 37.8 3.14 0.31720 [17]

37.01 143 46.5 3.861720 0.27 [17]

79.21 116 67.6 119.9 1.7 1.8 0.62 [16]

67.6 151.7 2.3153 2.2 0.49 [16]270 67.6 224.1 3.9 3.3 0.34 [16]

121.3 195.1 1.6198 1.6 0.78 [16]

99.9 195.1 1.9198 1.9 0.64 [16]

198 67.6 186.2 2.9 2.8 0.37 [16]

1.02 2.3 56.9 2.3 0.57 [18]57.9

1.2 2.3 38.647.4 1.9 0.56 [18]

1.7 2.3 22.638.6 1.4 0.57 [18]




Fig. 3. Fitting of the entrainment ratio for compression ratios higher

than 1.8.

wide range of compression, expansion and entrain-

ment ratios, especially those used in industrial appli-

cations. The developed correlations are simple and

very useful for design and rating calculations, since it

can be used to determine the entrainment ratio,

which, upon specification of the system load, can be

used to determine the motive steam flow rate and the

cross section areas of the ejector.

Acknowledgements

The authors would like to acknowledge funding

support of the Kuwait University Research Adminis-

tration, Project No. EC084 entitled ‘Multiple Effect

Evaporation and Absorption/Adsorption Heat

Pumps’.

Appendix A. Nomenclature

A cross section area (m2)

coef ficient of performance, dimensionlessCOP

Cr compression ratio defined as pressure of com-

pressed vapor to pressure of entrained vapor

Er expansion ratio defined as pressure of com-

pressed vapor to pressure of entrained vapor

m mass flow rate (kg/s)

M Mach number, ratio of fluid velocity to speed

of sound

M * critical Mach number, ratio of fluid velocity

to speed of sound

P pressure (kPa)P pressure drop (kPa)

universal gas constant (kJ/kg °C)R

Rs load ratio, mass flow rate of motive steam to

mass flow rate of entrained vapor

T temperature (K)

w entrainment ratio, mass flow rate of en-

trained vapor to mass flow rate of motive

steam

Greek symbols

compressibility ratio

ejector ef ficiency

Subscripts

locations inside the ejector1 – 7

b boiler

c condenser

diffuserd

e evaporator or entrained vapor

m mixing

n nozzle

p primary stream or motive steam

throat of the nozzlet

Fig. 4. Fitting of the entrainment ratio for compression ratios lowerthan 1.8.

dent of the nozzle and diffuser ef ficiencies, which

varies over a wide range, as shown in Table 2.

5. Conclusions

A semi-empirical model is developed for design and

performance evaluation of steam jet ejector. The

model includes correlations for the entrainment ratioin choked and non-choked flow, the motive steam

pressure at the nozzle outlet and the area ratios of

the ejector. The correlations for the entrainment ratio

are obtained by fitting against a large set of design

data and experimental measurements. In addition, the

correlations for the motive steam pressure at the noz-

zle outlet and the area ratios are obtained semi-em-

pirically by solving the mathematical model using the

design and experimental data for the entrainment ra-

tio and system pressures. The correlations cover a




Appendix B

B .1. Correlations of saturation pressure and temperature

The saturation temperature correlation is given by

T =

42.6776− 3892.7

(ln(P/1000)−9.48654)

−273.15

where P is in kPa and T is in °C. The above correlation

is valid for the calculated saturation temperature over a

pressure range of 10 – 1750 kPa. The percentage errors for

the calculated versus the steam table values are 0.1%.

The correlation for the water vapor saturation pressure

is given by

ln(P/Pc)

= T c

T +273.15−1

× 8

i =1

f i (0.01(T +273.15−338.15))(i −1)

where T c=647.286 K and Pc=22089 kPa and the values

of f i are given in the following table

f 3 f 1 f 4 f 2

−0.1155286−7.419242 0.0086856350.29721

f 7 f 8 f 6 f 5

0.002520658 −0.0005218680.001094098 −0.00439993

where P and T are in kPa and °C. The above correlation

is valid over a temperature range of 5 – 200 °C with apercentage error of 0.05% for the corresponding

values in the steam tables.

References

[1] E.E. Ludwig,AppliedProcess Design for Chemicaland Petrochem-

ical Plants, vol. 1, second ed, Gulf, Houston, TX, 1977.

[2] R.B. Power, Hydrocarb. Proc. 43 (1964) 138.

[3] H.T. El-Dessouky, H.M. Ettouney, Single effect thermal vapor

compression desalination process: thermal analysis, Heat Trans.

Eng. 20 (1999) 52 – 68.

[4] J.T. Munday, D.F. Bagster, A new ejector theory to steam jetrefrigeration, IEC 16 (1977) 442 – 449.

[5] H.J. Henzler, Design of ejectors for single-phase material systems,

Ger. Chem. Eng. 6 (1983) 292 – 300.

[6] J.H. Keenan, E.P. Neumann, A simple air ejector, J. Appl. Mech.

64 (1942) 85 – 91.

[7] D.W.Sun, I.W. Eames, Recent developments in the design theories

and applications of ejectors — a review, J. Inst. Energy 68 (1995)

65 – 79.

[8] I.W. Eames, S. Aphornaratana, H. Haider, A theoretical and

experimental study of a small-scale steam jet refrigerator, Int. J.

Refrig. 18 (1995) 378 – 385.

[9] D.W. Sun, I.W. Eames, Performance characteristics of HCFC-123

ejector refrigeration cycle, Int. J. Energy Res. 20 (1996) 871 – 885.

[10] N.H. Aly, A. Karmeldin, M.M. Shamloul, Modelling and simula-

tion of steam jet ejectors, Desalination 123 (1999) 1 – 8.

[11] B.J. Huang,J.M. Chang,C.P. Wang, V.A.Petrenko,A 1-Danalysis

of ejector performance, Int. J. Refrig. 22 (1999) 354 – 364.

[12] R.B. Power, Predicting unstable-mode performance of a steam jet

ejector, Am. Soc. Mech. Eng. FSPI 1 (1994) 11 – 15.

[13] S. Aphornratana, I.W. Eames, A small capacity steam-ejector

refrigerator: experimental investigation of a system using ejector

with moveable primary nozzle, Int. J. Refrig. 20 (1997) 352 – 358.

[14] D.F. Bagster, J.D. Bresnahan, An examination of the two-stream

theory of steam-jetejectors,Proceedingsof the EleventhAustralian

Conference on Chemical Engineering, Brisbane, 4 – 7 September,

1983.

[15] D. Sun,Variable geometry ejectors and theirapplications in ejector

refrigeration systems, Energy 21 (1996) 919 – 929.

[16] Y.M. Chen, C.Y. Sun, Experimental study of the performance

characteristics of a steam-ejector refrigeration system, Exper.

Therm. Fluid Sci. 15 (1997) 384 – 394.

[17] H.G. Arnold, W.R. Huntley, H. Perez-Blanco, Steam ejector as

an industrial heat pump, ASHRAE Trans. 88 (1982) 845 – 857.

[18] P. Everitt, S.B. Riffat, Steam jet ejector system for vehicle air

conditioning, Int. J. Amb. Energy 20 (1999) 14 – 20.

[19] N. Al-Khalidy, Performance of solar refrigerant ejector refrigerat-

ing machine, ASHRAE Trans. 103 (1997) 56 – 64.[20] S.L. Chen, J.Y. Yen, M.C. Huang, An experimental investigation

of ejectorperformance based upondifferent refrigerants,ASHRAE

Trans. 104 (1998) 153 – 160.

[21] M. Sokolov, D. Hershgal, Enhanced ejector refrigeration cycles

powered by low grade heat. Part 1. Systems characterization, Int.

J. Refrig. 13 (1990a) 351 – 356.

[22] M. Sokolov, D. Hershgal, Enhanced ejector refrigeration cycles

poweredby lowgradeheat. Part2. Design procedures, Int.J. Refrig.

13 (1990b) 357 – 363.

[23] N. Al-Khalidy, A. Zayonia, Design and experimental investigation

of an ejector in an air-conditioning and refrigeration system,

ASHRAE Trans. 101 (1995) 383 – 391.

[24] F.C. Chen, C.T. Hsu, Performance of ejector heat pumps, Energy

Res. 11 (1987) 289 – 300.

[25] M. Sokolov,D. Hershgal,Optimal couplingand feasibility of solarpowered year-round ejector air conditioner, Sol.Energy 50 (1993a)

507 – 516.

[26] M. Sokolov, D. Hershgal, Solar-powered compression-enhanced

ejector air conditioner, Sol. Energy 51 (1993b) 183 – 194.

[27] P. Menegay, A.A. Kornhauser, Ejector expansion refrigeration

cycle with underexpanded motive nozzle, Am. Inst. Aeronaut.

Astronaut. 2 (1994) 915 – 920.

[28] H.K. Abdel-Aal, A.S. Al-Zakri, M.E. El-Sarha, M.E. El-Swify,

G.M. Assassa, Other options of mass and energy input for steam

jet refrigeration systems, Chem. Eng. J. 45 (1990) 99 – 110.

[29] G. Grazzini, A. Mariani, A simple program to design a multi-stage

jet-pump for refrigeration cycles, Energy Convers. Manag. 39

(1998) 1827 – 1834.

[30] M.C. Huang, S.L. Chen, An experimental investigation of ejector

performance characteristics in a jet refrigeration system, J. Chin.Inst. Chem. Eng. 27 (1996) 91 – 100.

[31] M.L. Tomasek, R. Radermacher, Analysis of a domestic refriger-

ator cycle with an ejector, ASHRAE Trans. 1 (1995) 1431 – 1438.

[32] E.D. Rogdakis, G.K. Alexis, Design and parametric investigation

of an ejector in an air-conditioning system, Appl. Therm. Eng. 20

(2000) 213 – 226.

[33] D.W. Sun, Comparative study of the performance of an ejector

refrigeration cycle operating with various refrigerants, Energy

Convers. Manag. 40 (1999) 873 – 884.

[34] S.K. Gupta, R.P. Singh, R.S. Dixit, A comparative parametric

study of two theoretical models of a single-stage single-fluid, steam

jet ejector, Chem. Eng. J. 18 (1979) 81 – 85.



Ejector systems for fats,

oils, oleochemicals

Essential processes in the production of

natural fats and oils and derivative

oleochemicals are performed under

vacuum, i.e., at a pressure below

atmospheric. Such processes, including

solvent extraction, degumming, bleaching,

interesterification, fractionation,

winterization and deodorization, are

supported by ejector systems (Figure 1.).

Ejector systems are employed to produce

and maintain proper vacuum. The

complexity of the various processes

necessitates an integrated ejector system

for an optimized unit operation. An

integrated system will ensure that a proper

balance of operating and evaluated cost is

maintained while satisfying demands of

the process itself. Even though ejector

systems are an integral part of the

process, many users and operators of

these systems do not understand their

operational characteristics or what

influences their performance.

Ejectors

An ejector is a static piece of equipment

with no moving parts (Figure 2). The majo

components of an ejector are the motive

nozzle, motive chest, suction chamber, and

diffuser. An ejector converts pressure

energy of motive steam into velocity

Thermodynamically, high velocity is

achieved through adiabatic expansion of

motive steam through a conver-

Figure 1. Ejector System for soybean oil deodorizer

This article was prepared by J. R.

Lines, Vice President of Marketing for

Graham Corporation, 20 Florence

Ave., Batavia, NY 14020



gent/divergent steam nozzle. This

expansion of steam from the motive

pressure to the suction fluid operating

pressure results in supersonic

velocities at the exit of the steam

nozzle. Actually. the motive steam

expands to a pressure below the

suction fluid pressure. This creates the

driving force to bring suction fluid into

an ejector. Typically, velocity exiting amotive steam nozzle is in the range o

3,0004,000 ft./s.

High-velocity motive steam entrains

and mixes with the suction fluid. The

resultant mixture is still supersonic. As

the mixture passes through the

convergent, throat, and divergen

sections of a diffuser, high velocity is

converted back to pressure. The

convergent section of a diffuse

reduces velocity of the supersonic

flow as cross-sectional area is

reduced. This statement may appear tocontradict intuition but a

thermodynamic characteristic of gases

at supersonic conditions is tha

velocity is decreased as cross

sectional area is reduced. The diffuse

throat is designed to create a shock

wave. It is the shock wave that

produces a dramatic increase in

pressure as the flow goes from

supersonic to subsonic across the

shock wave. In the divergent section

of the diffuser, cross-sectional flow

area is increased and subsonic

velocity further reduced and

converted to pressure.

Ejector performance is summarized on

a performance curve (Figure 3). A

performance curve describes how a

given ejector will perform as a function

of water vapor equivalent loading

Other important information noted on

an ejector performance curve is the

minimum motive steam pressure

maximum permissible steam

temperature, and maximum discharge pressure (MDP).

Equivalent load is used to represent a

process stream, which may be made up

of many different components, such as

air, water vapor, free fatty acids (FFA

or various organics, in terms of an

equivalent amount of water vapo

(Figures 4,5). Heat Exchange Institute

(Cleveland, Ohio) Standards for Steam

Jet Ejectors describe the method used

to convert to water vapor-equivalen



or an air equivalent load. Water vapor

equivalent loading is often selected

because most factory performance

testing of an ejector is done with a

water vapor load (Table 1).

The performance curve may be used intwo ways. First, if suction pressure is

known for an ejector, the equivalent

water vapor load it handles is easily

determined. Second, if the loading to

an ejector is known, then it is possible

to estimate the expected suction

pressure for the ejector. If field

measurements differ from a

performance curve, then there may be

a problem with either the process,

utilities, or the ejector itself.

Condensers

Condensers may be categorized as

direct contact or surface type. Here we

will focus solely on surface-type

condensers, otherwise known as shell-

and-tube condensers. Direct-contact

condensers are still in use but because

of pollution concerns, they are not

often currently specified.

Condensers are manufactured in three

basic configurations: fixed tubesheet,

“U” tube, or floating head bundle

(Figure 6). The basic configurations

differ only in ease of maintenance andcapital cost, but thermodynamically

will perform similarly.

The primary purpose of a condenser in

an ejector system is to reduce the

amount of vapor load that a

downstream ejector must handle. This

will greatly improve the efficiency of

an ejector system. Although vacuum

condensers are constructed like

process shell-and-tube heat

exchangers, their internal design

differs significantly owing to the

presence of two-phase flow,

noncondensible gas, and vacuum

operation.

Vacuum condensers for fats, oils, andoleochemical applications generally

have the cooling water running

through the tubes. Condensation of

water vapor and organics takes place

on the shell-side the outside surface

area of the tubes. Generally, the inlet

stream enters through the top of the

condenser. Once the inlet stream

enters the shell, it spreads out along

the shell and penetrates the tube

bundle. A major portion of the

condensibles contained in the inlet

stream will change phase from vapor to

liquid. The liquid falls by gravity, runs

out of the bottom of the condenser

and down the tail leg. The remainder of

the condensibles and the

noncondensible gases are collected

and removed from the condense

through a vapor outlet connection. An

exception to the general rule is the firs

intercondenser of a deodorizer ejecto

system, where process vapors are on

the tube-side the inside surface of the

tubes.There are two basic types of vacuum

condensers typically offered. Fo

larger units approximately 30” in

diameter and larger a long air-baffle

design is used. A long air-baffle runs

virtually the full length of the shell and

is sealed to the shell to preven

bypassing of the inlet stream directly

to the vapor outlet. This forces vapors

to go through the entire tube bundle

before exiting at the vapor outlet

Similarly, smaller units use an up-and

over baffle arrangement to maximize

vapor distribution in the bundle. In

this configuration, the exiting vapo

leaves the condenser at one end only

The vapors are forced through a series

of baffles in order to reach the vapo

outlet.

As mentioned previously, a condense

is designed to limit the load to a

downstream ejector. In many cases

the inlet load to a condenser is many

times greater than the load to a

downstream ejector. Consequentlyany loss in condenser performance wil

have a dramatic effect on a

downstream ejector. This makes



the performance of an ejector extremely

dependent on the upstream condenser.

Inter and aftercondensers of an ejector

system are designed to condense steam

and condensible organics and coal

noncondensible gases (Figure 7). This

condensation will occur at a pressure

corresponding to the discharge pressure

of a preceding ejector and the suction

pressure of a downstream ejector.

Intercondensers are positioned between

two ejector stages and must operate

satisfactorily in order for the entire system

to perform correctly.

Precondensers

A precondenser, which is positioned

ahead of an ejector system, is a highly

specialized condenser and should be

considered part of the ejector system. The

operating pressure of a precondenser in

fats and oils processing is typically 10 mm

Hg absolute (abs) or less.

Process load from a distillation column or

still consists of large quantities ofcondensible vapors, such as glycerin

methyl esters or fatty alcohols, plus

noncondensiblegases. The low pressure

condition will result in extremely high

volumetric flow rates. It becomes a

challenge to effectively manage a large

volumetric flow rate at low pressure drop

while still accomplishing necessary heat

transfer. The tube field layout and

shellside baffling are quite special and

often unique to each application.

The tube pitch may be variable, with an

open pitch at the inlet and tighter pitchesat the outlet where volumetric flow is

considerably less than at the inlet

conditions. Location of a precondenser is

important for an optimized system. It is

key to locate a precondenser as close as

possible to the process vessel.

Attachment of a precondenser directly to

the vacuum vessel is preferred. This will

minimize pressure loss so as to reduce

utility consumption and maximize

condensation. Note that a precondenser is

part of an ejector system. Often specifiers

and purchasers separate a precondenser

from the ejector system. This will result in

more costly systems, with increased

operating costs. When properly designed

and integrated in an ejector system,

precondenser performance is optimized to

match the performance characteristics o

the ejector systems. The following

example highlights the importance o

maintaining lower pressure drop across a

precondenser (Table 2). As pressure drop

increases, condensation decreases.

UtilitiesMotive steam pressure, quality, and

temperature are critical variables. Cooling

water flow rate and inlet temperature are

important as well. Often, actual utility

supply conditions differ from those used



to design an ejector system. When this

occurs, system performance may or may

not be affected.

Steam

Motive steam supply condition is one of

the most important variables affecting

ejector operation. If motive supply

pressure falls below design pressure, then

the motive nozzle will pass less steam. If

this occurs, an ejector is not provided with

sufficient energy to entrain and compress

a suction load to the design discharge

pressure of the ejector. Similarly, if’ motive

steam supply temperature is appreciably

above the design value. then again,

insufficient steam passes through the

motive nozzle. With either lower than

design steam pressure or higher than

design steam temperature. the specific

volume of the motive steam is increased

and less steam will pass through a motive

nozzle. Less steam passing through a

motive nozzle results in less energy

available to do the necessary work (Table

3).

Any ejector may operate unstably if it is

not supplied with sufficient energy to

entrain and compress a suction load to thedesign discharge pressure. In certain

cases, it is possible to rebore an ejector

motive nozzle to a larger diameter if actual

supply steam pressure is below design or

its temperature above design. This larger

steam nozzle will permit the passage of

more steam through the nozzle, thereby

increasing the energy available to entrain

and compress the suction load.

If motive steam pressure is greater than

20% above design steam pressure, then

too much steam expands across the

nozzle. This has a tendency to choke the

diffuser throat of an ejector. When this

occurs, less suction load is handled by an

ejector and vacuum vessel pressure will

rise. If an increase in vessel pressure is

undesirable, then new ejector nozzles with

smaller throat diameters are required.

Steam quality is important. Any ejector is

designed to operate with dry steam

conditions. Wet steam is damaging to an

ejector system. Moisture droplets in

motive steam lines are rapidly accelerated

as steam expands across a motive nozzle.

High-velocity moisture droplets are

erosive. Moisture in motive steam lines is

noticeable when inspecting ejector

nozzles. The rapidly accelerated moisture

droplets erode nozzle internals. There is an

etched striated pattern on the diverging

section of a motive nozzle, and the nozzle

mouth may actually have signs of wear.

Also, the inlet diffuser section of an

ejector will show signs of erosion due todirect impingement of moisture droplets. It

is also possible to measure the exhaust

temperature from the ejector to determine

if wet steam conditions are present.

Typical ejector exhaust temperatures are in

the range of 250-300°F. If moisture is

present, a substantially lower ejector

exhaust temperature will exist.

To solve wet steam problems, all lines up

to an ejector should be well insulated. A

steam separator and trap should be

installed immediately before the motive

steam inlet connection of each ejector.

It is possible to have performance

problems due to wet steam. When

moisture droplets pass through an ejecto

nozzle, they decrease the energy available

for compression. This will reduce the

suction load-handling capability of an

ejector. Also, the moisture droplets may

vaporize within the diffuser section of the

ejector. Upon vaporization, the volumetric

flow rate within the ejector will increase

Here again, this reduces the suction load-

handling capability of an ejector. It is

recommended that supply steam be dry or

above 99% quality. With extremely wet

steam, any ejector will perform poorly.

Water

When cooling water supply temperature

rises above the design, ejector system

performance is penalized. A rise in cooling

water temperature lowers the available log

mean temperature difference (LMTD) of acondenser. Should this occur, that

condenser will not condense enough

steam or condensible organics, and

therefore there will be an increased vapor

load to a downstream ejector. Because of

inadequate condensation







charge pressure rises above design, an

ejector will not have enough energy to

reach that higher pressure. When this

occurs the ejector breaks operation and

there is an increase in vacuum vessel

pressure. When back pressure is above

design, possible corrective actions are to

lower the system back pressure, rebore the

steam nozzle to permit the use of moremotive steam that enables the ejector to

discharge to a higher pressure, or install

completely new ejectors. System back

pressure is the most common cause of

inadequate vacuum. Failing to make

adequate allowance for the back pressure

due to the pressure drop in the vent line or

tail leg, for the submergence of the tail leg

in a condensate receiver, or for site

barometric pressure will negatively affect

system performance.

Some ejector and condenser problems,

their effects, and possible corrective

actions are shown in Table 4.

Glycerin plants

Glycerin production is done at an

extremely high vacuum, very low absolute

pressure. Typically the operating pressure

of a glycerin vacuum flash still is below 10

mm Hg abs. Overhead load from the flash

still consists of glycerin, water vapor, and

air at temperatures approaching 400°F. In

one glycerin process, different glycerin product qualities are produced via

fractional condensation. Overhead

glycerin vapors from the vacuum flash still

are fractionally condensed by three

vacuum precondensers ahead of a four-

stage ejector system (Figure 8). The three

glycerin condensates produced by

fractional condensation have varied

commercial value.

The primary vacuum precondenser

fractionally condenses overhead load so

as to produce “commercially pure”

glycerin. Tight control of the

condensation profile is necessary to

maintain high purity levels. To maintain

control of product quality, vaporizable

water on the condenser tubeside is used

By controlling tubeside operating

pressure, the boiling temperature is varied

to maintain the outlet vapor temperature of

the condensing glycerin above the point

where impurities began to condense

thereby ensuring contaminant free

condensate.The secondary precondenser uses water

vaporization as the cooling medium as

well; however, the operating pressure o

the tubeside is lower. This condenser

produces glycerin condensate marketed as

“high gravity.” Again, the outlet vapor

temperature of the glycerin is maintained

so as to limit impurities in the condensate.

The final precondenser makes use of

tower water to condense and recover

remaining glycerin vapors exiting the

secondary condenser. The condensate is

recycled back to the process.

With three precondensers in series

operating at such low absolute



pressure, pressure drop across each

precondenser is extremely important. High

differential pressure drop not only results

in added utilities necessary for the ejector

system which backs up the condensers

but also reduces the amount of glycerin

recovered. The highest value

“commercially pure” glycerin production

is reduced when pressure drop is high,

Furthermore, high pressure drop increases

glycerin carryover to the ejector system

and as a consequence, increases product

loss.

Glycerin plant condensers often have

open tube pitches and large distribution

areas above and through the tube field.

Typical spacing between tubes in a

general heat exchanger would be 1.25

times the tube diameter. In vacuum

condensers operating at the low pressures

necessary to support glycerin production,spacing between tubes increases to 1.5 to

2.0 times tube diameter. This is necessary

to enable vapors to distribute above the

tube field and flow through the tube

bundle at velocities suitable for low

pressure drop, Target pressure drop is 1O

- l5% of the operating pressure.

Boiling water vacuum condensers are

rather sophisticated. The thermal and

hydraulic design warrants careful

consideration. To enable an optimized

design to be achieved, the precondenser

requirements should be discussed with

the ejector system manufacturer, Often

manufacturers with experience have

proprietary designs for this type of

service.

The foregoing is typical of one glycerin

process. Another process utilizes a

packed column with direct condensation

inside the column and a water-cooled

precondenser after the column for

reclamation of remaining glycerin.

Edible oil plantsEdible oil deodorization is done under

vacuum at very low absolute pressures.

Early systems operated at 5 to 6 mm Hgabs and had direct-contact condensers.

Today’s plants operate at 1.5 to 3 mm Hg

abs and have surface-type

intercondensers. This lower operating

pressure reduces stripping steam

consumption within the deodorizer, and

energy consumption is lower. Stripping

steam is used within the deodorizer to

lower fatty acid partial pressure, thereby

allowing the fatty acid to vaporize from the

oil. Therefore, the deodorizer overhead

load to the vacuum system is steam, free

fatty acid, fatty matter, volatile organic

compounds, and air. Normally, twoejectors in series compress deodorizer

overhead load to the first intercondenser.

Fatty acids solidify upon contact with

cold surfaces. The first intercondenser is

designed to handle fatty acid loading

without special provisions, the fatty acid

would rapidly solidify in the condenser

This first intercondenser is designed for

tubeside vacuum condensation, with

cooling water on the shellside. The fatty

acid solidified as it contacts the cold

surface of the tubesheet and tubes. If

provisions for removing solidified fatty

acid are not included, tube holes in the

tubesheet will plug. This reduces

performance and ultimately results in a rise

in deodorizer operating pressure. An

increase in deodorizer operating pressure

reduces the amount of fatty acid remova

from the oil; less will vaporize due to a

higher operating pressure. This degrades

product quality and marketability of the

oil.

The top head of the first intercondenser

has a nozzle that sprays caustic flushsolution on the inlet tubesheet to remove

fatty acid deposits (Figure 9). This is a

continuous washing operation, as fatty

acid buildup is rapid. Must of the fatty

acid is removed in the first intercondenser

and secondary condensers do not require

this feature.

An interesting concept that offered

appreciable savings in operating costs

was employed at an edible oil refinery in

Canada. In regions where cooling water

temperature varies significantly between

summer and winter months, it is possibleto control motive steam consumption to

optimize operating costs. In any

deodorizer ejector system,



the second stage ejector uses most of the motive steam required by

the ejector system. Steam consumption for this ejector may be

controlled as a function of cooling water temperature.

The principle at work in this arrangement is that as cooling water

supply temperature decreases, the operating pressure of the firs

intercondenser decreases as well. This occurs because colder cooling

water will increase the available LMTD, thus enabling that condenser

to operate at a lower pressure. As operating pressure of the firs

intercondenser is reduced, less energy is required to entrain andcompress the second stage ejector load to the operating pressure of

the condenser. A savings in motive steam usage is possible due to a

reduction in actual discharge pressure for the second stage ejector

(Figure 10).

An exacting test procedure must be followed by the ejecto

manufacturer to assess operating characteristics of the second-stage

ejector as a function of motive steam supply pressure. Motive steam

supply pressure to the second ejector is reduced as cooling water inlet

temperature is below design, Actually if water temperature is cold

enough, the second-stage ejector may be bypassed entirely, thus

tremendous savings in steam consumption may be realized during

winter months. It is also important to design the secondary equipment

those items downstream of the first intercondenser to follow the

performance of the first intercondenser. A caveat to bear in mind is

that processing of certain oils may result in increased fatty acid

fouling in the first intercondenser when cooling water is permitted to

drop below 75-80°F. Common operating practice is to control cooling

tower fan speed so as not to permit water temperature falling below

75°F.



Fatty alcohols/methyl estersFatty alcohol and methyl ester distillation plants will

use precondensers and three and four-stage ejector

systems. Once again, the precondenser should be

married to the ejector system. Operating pressure of

the distillation column is less than 10 mm Hg and will

have 10,000 to 30,000 pounds per hour (pph) Cl2 load

or greater. A precondenser should be mounted

directly atop the vacuum column, as shown in Figure

11. This keeps pressure drop to a minimum but will

require a special layout for optimal performance.

Either tempered water or boiling water is used on the

tubeside to effect organic condensation on the

shellside of the condenser. Here the temperature of

the tubeside fluid is important so as to maintain the

metal temperature above the point where methyl

esters will solidify. An added benefit from boiling

water is that the large enthalpy change associated

with boiling water permits less water to be used as

opposed to the amount required if tempered water isused. The figure depicts a horizontal condenser

mounted directly on the distillation column, which is

typical of tempered water-cooled precondensers.

SummaryComplexity of ejector systems in fats, oils, and

oleochemical production requires that careful

consideration be given to their design, installation,

and performance troubleshooting. An ejector system

is truly an integral part of the process. If properly

designed, an ejector system will provide problem free

performance. When precondensers are involved, it is

important to integrate the precondenser into theejector system design. This will ensure a unitized

design that minimizes capital cost and operating

expenses.



LESSONS FROM THE FIELD

- EJECTOR SYSTEMSJames R. Lines, Graham Corporation,

USA, presents the problems associatedwith ejector system performance and

subsequent solutions.

Figure 1. Ejector cross-sectional drawing

H ydrocarbon Engineering has previously reported on ejector system

fundamentals, operating characteristics, and guides for

troubleshooting1. Moving on from that stage, the current article provides

real world ejector system performance limitations uncovered during

routine performance surveys. Corrective action undertaken to improve

performance is documented and discussed in detail. Principles from theinitial article are used as the tools to define the cause of a particular

limitation and the eventual solution. It should be noted that the corrective

actions described were unique to the particular problems discussed. It

will not always be possible to apply the same procedure to a

comparable performance problem. A review of general corrective

techniques is discussed where applicable. Ejector system

manufacturers should be consulted as a first course of action, and

guide fixes are often possible.

Survey 1 - nylon intermediate production

facilityNitrogen gas bleed for pressure control A North American petrochemical company manufacturing nylon

intermediates was operating a vacuum flasher supported by a

precondenser and two stage ejector system. Overhead load from the

vacuum flasher consisted of 160 000 pph (72 600 kg/hr) of mixed

nitriles at a pressure of approximately 35 torr.

The precondenser produced adequate vacuum, but the two stage

ejector system that extracted non-condensibles from the precondenser

was performing in an unstable manner. Suction pressure of the first

stage ejector was cycling between the design 35 torr and up to as high

as 75 - 80 torr.

Vacuum flasher pressure was unaffected by the ejector instability,

however, plant personnel had concerns that poor ejector performance

may at some point have a negative impact on vacuum flasher operating

pressure.

Both precondenser and vacuum system were supplied by the ejecto

system manufacturer. The manufacturer dispatched a service enginee

to the site to survey the equipment and its performance. Figure

depicts the pressure profile of the equipment.

The service engineer initially inspected vapor piping and condensate

drain legs to ensure equipment layout was satisfactory. Attention was

then focused on the utilities. Motive steam pressure was measured a

the inlet to each ejector, and actual motive steam supply pressure to the

ejectors was 140 psig (9.7 barg). The ejector motive steam nozzles

were designed to pass the required steam at 125 psig (8.6 barg)

Although the motive steam pressure was above design and

consequently, more steam was being consumed by the ejectors, the

excessive steam consumption was not enough to cause poo

performance.

The cooling water inlet temperature to the condensers was below

design, and temperature rise across each condenser was less than

the design. Inlet cooling water was designed for 89.6 °F (32 °C) and thewater flowed in series from the first intercondenser to the

aftercondenser. The actual inlet water was at 85 °F (29.4 °C). The tota

temperature rise across both condensers at design was 29 °F (16.1

°C). The actual temperature rise was 13 °F (7.2 °C). The lowe

temperature rise would suggest greater cooling water usage or lowe

condensible vapor discharge from the precondenser, neither of which

would cause poor ejector system performance.

An ejector system experiencing unstable suction pressure is typically

operating in a broken mode. Broken ejector performance is often

caused by low motive steam pressure, which has already been ruled

out, a fouled intercondenser, high cooling water temperature or wate

flow, both of which have been ruled out, non-condensible loading.

While inspecting the ejector system, the service engineer noticed periodic audible change in ejector operation. This audible change plus

an unstable suction and discharge pressure first stage ejecto

confirmed that this particular ejector was the trouble

The service engineer noticed plant personnel had installed a

pneumatically controlled control valve that bled nitrogen to the suction

of the first stage ejector. Plant personnel installed a nitrogen bleed as a

means of controlling suction pressure to allow the vacuum flasher to

operate at a consistent pressure even at reduced charge rates

Pressure in the top of the vacuum flasher was sensed and a signa

sent to the control valve to bleed nitrogen to the first stage ejector if the

Figure 2. Precondenser to left of vacuum flasher



vacuum flasher pressure fell below design. Bleeding nitrogen, which is

non-condensible, to the suction of a multi-stage condensing ejector

system will result in unstable performance.

An ejector system is designed to handle non-condensible loading

associated with the process. Ejectors downstream of the first

intercondenser are designed to handle process related non-

condensibles and associated vapors of saturation. Bleeding in nitrogen

to act as an artificial load for the first stage ejector and to elevate

suction pressure resulted in non-condensible overloading of the

downstream ejector, which is the ejector that is downstream of the

first intercondenser.

34 torr

40 torr

33-75 torr

36 torr

N2 Supply

2nd

Stage Ejector

Af te rc ond en ser

1 st Intercondenser

1s Stage Ejector

Act ua l V alu es

Design Values

PI C

Precondenser

Vacuum Flasher

279 torr

240 torr 200-262 torr

228 torr

773 torr

940 torr

Figure 3. Survey 1 pressure profile

8-13 torr

10 torr

50-70 torr

65 torr

3rd Stage Ejector

2nd

Stage Ejector

Aftercon denser

2nd Intercondenser

1st

Intercondenser

1 st Stage Ejector

Actual Value s

Design Values

VacuumDistillation

Unit

46-60 torr

62 torr

230 torr

215 torr

800 torr

Figure 4. Survey 2 pressure profile

Once the first stage ejector began to handle more non-condensible

loading than it was designed for, the down-stream ejector could not

handle that increased non-condensibles, plus the proportionate

increase in vapors of saturation, at the achievable discharge pressure

of the first stage ejector. This discontinuity in the achievable discharge

pressure of the first stage ejector and suction pressure maintained by

the second stage ejector based on higher non-condensible loading

resulted in the first stage ejector breaking operation.

The service engineer instructed plant personnel to dis-assemble the

nitrogen bleed arrangement and to install recycle control piping around

the first stage ejector. For any multi-stage condensing ejector system

the preferred way to maintain performance and suction pressure is to

recycle discharge from an ejector immediately preceding the firs

intercondenser back to the suction of the system. In this way, non

condensible loading is never allowed to increase above design, thus

ensuring broken ejector operation will not occur. Again, vacuum flashe

pressure is sensed and a signal sent to the recycle control valve

which will modulate and permit the recycle of vapor flow back to the

suction of the first stage ejector. Once the plant installed this form of

recycle control, stable ejector operation was maintained.

A caveat for this correction is that the most practical method o

controlling operating pressure of a precondenser/ejector system is to

control cooling water flowrate. Cooling water flowrate may be reducedwhen process charge rate is below design. By lowering wate

flowrate, the water temperature rise across the precondenser wil

increase, which has the effect of lowering the Imtd. Controlling lmtd wil

control operating pressure of the precondenser.

The recycle control arrangement suggested and used to correct firs

stage ejector instability will not work if the operating pressure of a

precondenser permits condensation of steam. The composition of

recycle flow around an ejector consists of non-condensibles plus

steam. As the recycle flow is brought around to the suction of the firs

stage ejector, the recycled steam will be drawn to the precondenser if

the operating pressure permits condensation of steam. When this

occurs and recycled flow goes to the precondenser rather than

through the first stage ejector, control of suction pressure is no

possible.

Survey 2 - West Coast fuels refinery

Improper replacement intercondenser A West Coast refiner was operating a fuels vacuum distillation unit tha

experienced erratic performance after replacing an intercondense

supplied by the original ejector system manufacturer with one designed

and built by a local heat exchanger fabrication shop. The as sold

system was designed to provide performance described in Figure 4

The service engineer had no prior knowledge that the user installed a

replacement intercondenser.

The first stage ejector was operating in a broken mode, with both

suction and discharge pressure remaining unstable. Furthermore

shellside pressure drop across the first intercondenser was almos

three times the design pressure drop.

Motive steam supply condition was approximately at the design value

so the service engineer ruled out inadequate steam pressure. High

pressure drop across the first intercondenser would suggest a fouling

problem, cooling water flowrate limitation, high inlet water temperature

high noncondensible loading, or excessive hydrocarbon loading. Prio

to detailing a method to determine the actual cause, the service

engineer discussed general performance characteristics with uni

operators. At that time, it was discovered that the first intercondense

was replaced.



2nd Stage Ejector

1st Stage Ejector

Turbine

Surface Condenser

CombinedInter/Aftercondenser

Act ual Val ues

Design Values

ExcessiveVapor Plume

113 torr

75 torr

250 torr

156 torr

113 torr

50 torr

Figure 5. Survey 3 pressure profile.

24-25 torr

10 torr

114-124 torr

100 torr

3rd

Stage Ejector

2nd

Stage Ejector

∆P Tubeside

Aft erc ond ens er

2nd Intercondenser

1st Intercondenser Isolated

1 st Intercondenser

1st Stage Ejector

Act ual Val ues

Design Values

VacuumDistillation

Unit

105-115 torr

105 torr

248-252 torr

249-253 torr

864 torr

95 torr

292 torr

104 torr

280 torr

890 torr

105-115 torr

95 torr

86 torr

90 torr 25 psi

5 psi

114-124 torr

100 torr

Figure 6. Pressure and temperature profile.

Upon visual inspection of the installed unit and its name-plate, the

service engineer realized it was the design of another vendor. Thatvendor did match the original intercondenser’s tube count and external

dimensions, but after a thorough review of fabrication drawings, it was

evident the vendor failed to design the shellside baffling properly to

manage hydraulic and thermal requirements. Vacuum condensers have

special shell side baffling to ensure minimal pressure drop, non-

condensible gas cooling, and separation of non-condensibles and

condensate. It is typical to have different baffle spacing at strategic

locations within the shell of a vacuum condenser or to incorporate a

long air baffle design. The vendor who replaced the intercondenser

used conventional software to model the performance. This in turn

resulted in a design having fully baffled flow, and consequently

excessive pressure drop on the vapor side.

In this particular instance, high pressure drop across the shellside

caused the system to break performance. The first stage ejector could

not overcome the added pressure drop and reach a discharge

pressure where the second stage ejector would operate. This

discontinuity resulted in the first stage ejector breaking operation

which was characterized by unsteady suction pressure and back-

streaming of motive steam into the vacuum distillation tower. Both

performance conditions were unsatisfactory to the refiner.

Although the plant engineers were reluctant to accept that the

condenser was the problem, they did agree to install a new condense

designed by the ejector system manufacturer. Once the properly

designed condenser was installed and the system restarted

performance returned to a satisfactory level.

Survey 3 - Canadian ammonia/urea fertilizer

complex

An ammonia plant syngas compressor provided less than design

horsepower due to high back pressure from a condensing turbine

steam surface condenser. The turbine exhaust condenser maintained

113 torr back pressure, but based on the cooling water temperaturethe expected back pressure should have been 75 torr. A service

engineer was dispatched to the site to evaluate the steam surface

condenser and exhauster performance to determine the cause of the

elevated back pressure.

The steam surface condenser was supported by a two stage ejector

system condenser exhauster (Figure 5). The service engineer noticed

a substantial exhaust plume from the aftercondenser vent.

Normally, steam surface condenser and exhauster systems are

vacuum tight, with air inleakage less than Heat Exchange Institute

design values, with typical air inleakage of 5 Ibs/hr or less. An

excessive exhaust plume from an aftercondenser does suggest high

air inleakage. There was an air leakage meter installed on the vacuum

system, and when activated, the measurement was off the scale.

The service engineer elected to isolate the surface condenser from the

ejector system. By isolating the surface condenser, it would be

possible to determine if excessive air leakage was from the surface

condenser or upstream piping, or if it was within the exhauster itself

Once a surface condenser is isolated from a vacuum system and the

operating pressure of the condenser does not appreciably increase

over time, the air inleakage must be downstream of the surface

condenser.

The condenser was isolated from the vacuum system and pressure

stayed fairly constant. This confirmed the air inleakage was

downstream of the condenser and that it was in the exhauster system

A closer look at the installation determined that a l/4 in. instrumenconnection was left open and was not plugged. Evidently, a pressure

gauge was damaged and plant personnel removed it but failed to

replace it. The open connection permitted substantial quantities of air to

leak into the ejector system and cause poor operation. The condenser

was then brought on line once the connection was plugged and afte

the system was allowed to stabilize, steam surface condense

operating pressure reached 80 torr, which was in the range of wha

was expected. The syngas compressor returned to full power once

this correction was made.



Survey 4 - Gulf Coast refinery

Fouled intercondenser A Gulf Coast refiner was operating a damp crude

vacuum distillation tower that was designed for 10

torr tower top pressure but was maintaining only 24

-25 torr. The first stage ejector was surging and

back-streaming into the vacuum distillation unit. A

factory service engineer was dispatched to the site

to perform a system survey and evaluate causes of the poor performance.

Figure 6 documents as sold performance and what

was measured in the field.

Broken first stage ejector performance may be

caused by improper motive steam pressure, elevated

inlet cooling water temperature, lower than design

cooling water flowrate, a fouled first intercondenser, or poor operation

of a downstream ejector. The performance survey indicated motive

steam supply conditions were satisfactory. Cooling water temperature

rise and pressure drop across the first intercondenser suggested the

problem was here.

Design cooling water temperature rise across the first intercondenser was 14 °F (7.8 °C), however, the actual temperature rise was 19 °F

(10.6 °C). Possible causes for an elevated temperature rise would be

lower than designed cooling water flow or an increase in condensible

load to the condenser. Pressure drop across the tubeside of the con-

denser gave an indication that something was wrong. The actual

tubeside pressure drop was 25 psi (1.7 bar) while the design was only

5 psi (0.35 bar).

The tubeside of the condenser was fouled and the increased pressure

drop across the condenser caused the recirculating pumps to circulate

less water. Tubeside fouling to produce such an elevated pressure

drop would be severe and actual tube blockage must have occurred.

Tubeside fouling deterred heat transfer and did not permit proper

condensation of shell side vapors. This increased the pressure drop on

the shell side of the condenser and elevated its operating pressure. By

not permitting proper condensation of shellside vapors, the increased

outlet flow of vapors caused an increase in pressure drop.

The first stage ejector could not overcome the elevated shell side

pressure drop and, consequently, broke operation. The broken

operation resulted in unstable suction pressure, surging and back-

streaming of motive steam into the vacuum distillation unit. The first

intercondenser was pulled from the platform and taken down to grade.

At grade, the bundle was removed to inspect the shell side for fouling

and to rod out the tubes. The shell side did not experience excessive

fouling, but the tubeside had tubes blocked with solidified calcium

carbonate and other inverse solubility salts.

Once the tubeside was cleaned and returned to acceptable condition,

the bundle was reinstalled in the condenser, and the condenser taken

up to the vacuum unit for re-hook up. When the system was brought in

service, the tower top pressure was maintained at approximately 10

torr and system performance was stable.

Conclusion

Ejector systems provide extremely reliable performance, but they do

require periodic maintenance. It is recommended that routine surveys

be performed to document actual behavior and performance of the

ejector system. An ejector system may be performing at less than

optimal conditions for a variety of reasons

such as improper utilities, fouled

condensers, mechanical damage

excessive process load, excessive non

condensible load or improper installation.

A skilled vacuum technician, most often

from the ejector system manufacturer

should conduct the routine surveys and

issue performance reports. The

performance surveys may be conducted

on line without affecting the process. The

performance reports will document actua

performance at a point in time, discuss

corrective action where applicable and

offer preventative maintenance

suggestions.

If performance problems arise, the origina

supplier of the vacuum system should be

consulted. If necessary, a request should be made for a service

engineer to be dispatched to offer support on site. Actual corrective

action to take is situation dependent and requires a thorough

understanding of variables that influence ejector system performance.

References1 LINES J R and SMITH R T, Ejector system troubleshooting

Hydrocarbon Engineering, Part 1 January/February 1997 pp. 69 - 78

Part 2 March/April 1997 pp 35 - 40

Palladian Publications 1999

For More Information

Please Contact:

Graham Corporation

20 Florence Avenue

Batavia, New York 14020

USA

Telephone: 716 343 2216 Fax: 716 343 1097

Website: http/www.graham-mfg.com

Figure 7. First stage ejectors for

CVDU.





TECHNOLOGY

Understanding ejector systems necessary

to troubleshoot vacuum distillation James R. Lines Graham Corp. Batavia, NY

.

A complete understanding of ejector

system performance characteristics can

reduce the time and expense associated

with troubleshooting poor crude

vacuum distillation unit (CVDU)

performance.

Variables that may negatively impact

the ejector-system performance of

vacuum-crude distillation units include

utilities supply, corrosion and erosion

fouling, and process conditions.

Fig 1. Fig. 2



Tables 1 and 2 are troubleshooting guides to

ejector and condenser problems in vacuum

ejector systems. Fig. 1 is a photo of an

installed ejector at a CVDU.

Two actual case studies conducted by service

engineers on CVDU-ejector systems show

how to troubleshoot ejector problems. The

first problem was a result of improper

replacement of an intercondenser, and the

second was a result of underestimation of

noncondensible loading during design, which

has recently become a common problem.

EjectorsAn ejector converts pressure energy of

motive steam into velocity. It has no moving

parts. Major components of an ejector

consist of the

motive nozzle, motive chest, suction

chamber, and diffuser (Fig. 2).

High velocity is achieved through adiabatic

expansion of motive steam across a

convergent/divergent steam nozzle. This

expansion of steam from the motive pressure

to the suction fluid operating pressure results

in supersonic velocities at the exit of the

steam nozzle.

The motive steam actually expands to a

pressure below the suction fluid pressure.

This expansion creates a low-pressure region,

which draws suction fluid into an ejector.

Typically, velocity exiting a motive steam

nozzle is in the range of 3,000-4,000 fps. Thi

high-velocity motive steam then entrains and

mixes with the suction fluid. The resultanmixture is still supersonic. As the mixture

passes through the convergent, throat, and

divergent sections of a diffuser, high velocity

is converted back to pressure.

The convergent section of a diffuser reduce

velocity as cross sectional area is reduced

Intuitively, one normally thinks that as flow

area is reduced, velocity is increased. But

unique thermodynamic phenomenon occur

with gases at supersonic conditions: A

cross-sectional flow area is reduced, th

velocity is reduced.

The diffuser throat is designed to create shock wave. The shock wave produces

dramatic increase in pressure as the flow goes

from supersonic to subsonic across it. In the

divergent section of the diffuser, cross

sectional flow area is increased and velocity is

further reduced and converted to pressure. A

shock wave occurs in the diffuser throat when

the compression ratio of an ejector is 2:l o

greater, which is the case with CVDU ejecto

systems.

An ejector-performance curve gives th

expected suction pressure as a function o

water-vapor equivalent loading (Fig. 3). Hea

Exchange Institute Standards for Steam Je

Ejectors describes the method to convert the

mixture (air, water vapor, and variou

hydrocarbons) to a water-vapor equivalent o

an air-equivalent load.

Other important information noted on an

ejector performance curve includes the

minimum motive steam pressure, the

maximum motive steam temperature, and

Fig. 5





temperature is appreciably above the

design value, insufficient steam passes

through the motive nozzle. Both lower-

than-design steam pressure and higher-

than-design steam temperature increase

the specific volume of the motive steam

and reduces the amount of steam

through a motive nozzle.

In certain cases, it is possible to re-bore

an ejector-motive nozzle to permit the

passage of more steam through thenozzle, thereby increasing the energy

available to entrain and compress the

suction load.

If motive-steam pressure is more than

20% above design, too much steam

expands across the nozzle. This often

chokes the diffuser throat of an ejector.

When this occurs, less suction load is

handled by an ejector, and the CVD-

column pressure rises. If an increase in

column pressure is undesirable, then

new ejector nozzles with smaller throat

diameters are required.

Steam qualityWet steam is very damaging to an

ejector system because high-velocity

moisture droplets are erosive. These

droplets are rapidly accelerated as steam

expands across a motive nozzle.

Erosion of nozzle internals caused by

wet motive-steam is noticeable when

inspecting ejector nozzles or diffuser

internals. There is an etched striated

pattern on the diverging section of

motive nozzle, and the nozzle mouth

may actually wear out. Also, the inle

diffuser section of an ejector will show

signs of erosion as a result of direc

impingement of moisture droplets (Fig

4a).

Fig. 4b depicts an ejector cutaway

showing severe damage caused by we

steam. The inlet diffuser shows

a e

Table 3



substantial metal loss. Metal-scale buildup

can be seen in the outlet diffuser section.

The exhaust temperature from the ejector can

determine if the steam conditions are present.

Typical ejector exhaust temperatures are in

the range of 250 to 300° F. If moisture is

present, a substantially lower exhaust

temperature will exist.

To solve wet-steam problems, all lines up to

an ejector should be well insulated. A steamseparator and trap should be installed

immediately before the motive-steam inlet

connection of each ejector. In some instances,

a steam superheater may be required.

Wet steam can also cause performance

problems. Moisture droplets through an

ejector nozzle decrease the energy available

for compression. This reduces the suction-

load handling capacity of an ejector.

Also, the moisture droplets may vaporize

within the diffuser section of the ejector.

Upon vaporization, the volumetric flow rate

within the ejector increases. Here again, thisreduces the suction-load capacity of an

ejector.

Cooling water conditionsA rise in cooling-water temperature lowers

the available log mean temperature difference

(LMTD) of a condenser. Should this occur,

the condenser will not condense enough steam

and condensible hydrocarbons. This will

increase the vapor load to the downstream

ejector.As a result of inadequate condensation, there

also is an increase in pressure drop across the

condenser. If an ejector following this

condenser cannot handle an increased vapor

load at the operating pressure of a condenser,

the operating pressure of the condenser will

rise and the system will break performance.

Broken ejector system performance is

characterized by a higher-than-design CVDU

tower-top pressure. The tower-top pressure

may become unstable.

This may also occur if the cooling-water flow

rate is below design. At lower-than-design

flow rates, there is a greater water-

temperature rise across a condenser. Here

again, this will lower the available LMTD.

Poor performance is further exacerbated as a

result of a lower heat transfer coefficient

resulting from low-water flow rate.

Problems with cooling water normally occur

during summer months. During the summer,

the water is at its warmest, and demands on

refinery equipment are highest. If cooling-

water flow rate or temperature are off design,

new ejectors or condensers may be required

to provide satisfactory operation.

Corrosion and erosionCorrosion may occur in ejectors, condensers,

or Vacuum piping. Extreme corrosion may

cause holes and allow a system. Air leakage

into the vacuum system. Air leakage into a

vacuum system will deteriorate performance

and can result in broken ejector operation.

A common corrosion problem occurs when

carbon-steel tubing is used in condensers.

Although carbon steel may be suitable for the

crude feed-stock, it is not always the best

choice for an ejector system. Although carbon

steel has a lower capital cost, operating

problems can outweigh modest up-front

savings.

During extended periods of shutdowns for maintenance or revamps, a condenser with

carbon-steel tubing will be exposed to air,

oxidize, and develop a scale buildup. When an

ejector system starts up, this buildup can

severely foul the condensers and prevent

proper operation of the vacuum system.

Poor steam quality and high velocities may

also cause erosion of the diffuser and motive-

nozzle internals. Ejector manufacturers will

provide certified information that defines the

motive nozzle and diffuser throat diameters.

If a routine inspection of these parts indicates

an increase in cross sectional area over 7%,then performance may be compromised, and

replacement parts are necessary.

Threaded steam connections may experience a

phenomenon termed wire drawing, or wire

cutting. Loose threads provide a leak path for

the steam. Over time, the steam will destroy

the threaded joint or even put a hole in the

piece. A hole leads to a steam leak within the

ejector, which will act like a suction load,

thereby reducing the system’s performance.

FoulingIntercondensers and aftercondensers are

subject to fouling on both the tube side and

the shell side. Fouling deters heat transfer.

Cooling-tower water, often used as th

cooling fluid for vacuum condensers, is

normally on the tube side. Over a prolonged

period of time, actual fouling may exceed th

design value, and condenser performanc

becomes inadequate.

Vacuum-tower overhead gases, vapors, and

motive steam are normally on the shell side o

a condenser. Depending on fractionation and

the type of crude processed, a hydrocarbon

film may develop on the outside surface o

the tubing. This film deters heat transfer.

Fig. 5 illustrates how severely a condense

may be fouled. In this example, not only did

the tubing have a hydrocarbon film, bu

solidified hydrocarbon product adhered to th

tubing. The solidified material blocked th

flow, resulting in poor performance and an

elevated pressure drop.

When actual unit fouling exceeds design

values, a condenser performs inadequately

Once fouled, a condenser is unable t

condense sufficient quantities of hydrocarbon

vapors and motive steam. The result o

condenser fouling is an increase in vapor load

to a downstream ejector and an increase in

condenser-operating pressure. Ultimately,

preceding ejector will break operation.

Routine refinery procedures should includ

periodic cleaning of the tube side and the she

side of condenser bundles.

Process conditionsVacuum system performance may be affecte

by several process variables: non-condensibl

gas loading, condensible hydrocarbons, and

vacuum system back pressure.

Ejector systems are susceptible to poo

performance when noncondensible loadin

increases above design. Noncondensibl

loading to an ejector system can be caused by

air leakage into the system, the presence olight hydrocarbons, or the existence o

cracked gases from a fired heater.

The impact of higher-than-design

noncondensible loading is severe. A

noncondensible loading increases, the amoun

of saturated vapors discharging from

condenser increases proportionately.



The ejector following a condenser may

not be able to handle increased loading

at that operating pressure of the

condenser. The ejector preceding that

condenser is unable to compress to a

higher discharge pressure. This

discontinuity in pressure causes the

preceding ejector to break operation. I

When actual noncondensible loading is

consistently above design, new ejectorsare required. Depending on the severity

of noncondensible overloading, new

condensers may be required as well.

Recently, several CVDU revamps in the

U.S. Gulf Coast experienced startup

difficulties due to inaccurate estimates of

actual noncondensible loading.

As different crude oils are processed, or

as refinery operations change, the

composition and amount of condensible

hydrocarbons handled by an ejector

system vary. Condensable hydrocarbon

loading may become so much greaterthan design that condenser or ejector

performance is adversely affected.

Another possible affect of increased

condensible hydrocarbon loading is an

increased oil-condensate film on the

tubing, and consequently, a reduction in

the heat transfer rate. This situation may

result in increased vapor discharge from

a condenser. Unstable operation of the

entire ejector system may result. To

overcome this type of performance

limitation, new condensers or ejectors

may be required.

Vacuum system back pressure may have

an overwhelming influence on

satisfactory performance. If the actual

discharge pressure rises above design,

an ejector will not have enough energy

to reach that higher pressure. When this

occurs, the ejector breaks operation, and

there is an increase in CVDU tower-top

pressure.

When back pressure is above design,

possible corrective actions include

lowering the system back pressure,reboring the steam nozzle to permit the

use of more motive steam, or installing

new ejectors.

Case 1:

Improper intercondenser A West Coast refiner experienced erratic

system performance after replacing an

intercondenser supplied by the ejector

system manufacturer with one designed

and built by a local heat exchanger

fabrication shop. The ejector system

vendor dispatched a service engineer to

investigate the cause of the problem

without knowing about the replacement

intercondenser.

The actual performance of the system

differed from the “as sold” system (Fig.

6). The first-stage ejector was operating

in a broken mode with both suction and

discharge pressure remaining unstable.

Pressure drop across the firstintercondenser was excessive -at 8.5 mm

Hg instead of 3 mm Hg.

Broken first-stage ejector performance

and high-pressure drop across the first

intercondenser suggested one of the

following problems: fouling, cooling-

water flow rate limitation, high inlet

water temperature, or excessive

hydrocarbon loading.

Prior to detailing a method to determine

the actual cause, the service engineer

discussed general performance

characteristics with unit operators. Atthat time, he discovered that the first

intercondenser had been replaced by

another vendor.

The vendor had matched the original

unit’s tube count and external

dimensions, but failed to properly

design the shellside side baffling to

effectively manage hydraulic and

thermal requirements.

Vacuum condensers have special

shellside baffling to ensure minimal

pressure drop, noncondensible gas

cooling, and separation of

noncondensibles and condensate. It is

typical to have different baffle spacing at

strategic locations within the shell.

The vendor of the replacement

condenser used conventional software to

model the performance. The new

condenser design had a fully baffled

flow, and consequently a high-pressure

drop.

In this instance, the high-pressure drop

across the intercondenser caused the

system to break performance. The first-stage ejector could not overcome the

added pressure drop and reach a

discharge pressure in which the second-

stage ejector would operate.

Once the replacement unit was pulled

out and a properly designed condenser

put in, system performance was

satisfactory.

Case 2:

Underestimated loadingA U.S. Gulf Coast refiner grossly

underestimated its noncondensible

loading when it modernized a CVDU to

process sour South American crude. The

modernization effort involved an

entirely new ejector system.

Upon startup of the CVDU, the ejectosystem was not performing properly

Tower-top pressure was significantly

above design, and it was unstable.

Initial investigation verified utility

conditions. The ejector system was

designed for 140 psig motive steam, and

the actual supply pressure varied

between 138 and 144 psig.

Next, the cooling water was evaluated

Design inlet temperature was 88° F., and

the actual supply temperature was a

72.3° F. Temperature rise and pressure

drop across each condenser did nosuggest an abnormality. The equipmen

was new, so fouling was ruled out.

A detailed analysis of the sour South

American crude oil was in order.

The design and actual vacuum towe

overhead compositions are shown in

Table 3.

The actual simulation was too differen

from design conditions. Significan

equipment modifications were needed

to achieve the desired charge rate and

vacuum level.

The steam equivalent loads werecalculated to be about 17,500 lb/hr and

23,000 lb/hr for design and actua

loading, respectively. According to the

performance curve, at the higher load

the first-stage ejector would maintain

about 19 mm Hg absolute pressure in

lieu of the design 14 mm Hg. The refine

agreed to accept the higher pressure.

Because the noncondensible loading

values were drastically different (more

than twice as much as design) new

equipment was necessary.

The refiner added redundant ejector

and condensers after the firs

intercondensers to handle the additiona

noncondensible load. The system

stabilized after two parallel trains o

secondary equipment were installed

Tower-top pressure was still above

design but within an acceptable range.

Figs. 7a and 7b depict the “as sold”

performance and the revamped

operation.



Air Ejectors Cheaper Than Steam

When all the cost factors are considered, t

air-operated ejector often proves to be the superior meth

for producing vacuum. Here are figures you can u

F. Duncan Berkeley

For many years the air-operated ejector has been a neglected child in the field of

vacuum producing apparatus. It hasbeen greatly overshadowed by its highlysuccessful, fully reliable and popular kin,the steam ejector. The popularity of thesteam ejector has been somewhatustified because air-operated ejectorshave been limited in their use by arelatively expensive and somewhatscarce supply of high-pressure motiveair. Major reasons for selecting steamrather than air to operate ejectors have

been the unavailability of air compressors and the relatively high cost

of compressed air in most localities.

Improvements in air compressors havegreatly reduced the cost of compressedair as compared to 20 years ago; andthe greater availability of compressed air in process plants today makes the air ejector a reasonable and in someinstances a preferred means of producing a vacuum.

The fact that air is a non-condensgas under common conditions

temperature and pressure, limits its as a propelling material for ejectortwo or three stages. In a steam ejethe steam from each stage of multistunits can usually be condensed inintercondenser and the successstage need handle only the ncondensible gases plus a relativsmall saturation component from previous stages. By condensing the



motive steam from previous stages, it isboth economical and practical to use asmany as five or more stages.

CONSIDER ALL THE FACTORS

Recent tests and studies on air-operated ejectors have brought to lightsome rather interesting and useful factsconcerning these units. The results,although neither highly revolutionary nor startling, prove that the air jet has thesame desirable feature as the steam jet;and in some instances can prove to bevery economical and more desirablethan the steam jet.

All factors of cost should be carefullyconsidered for a specific application.They are:•Initial cost of the equipment used to

produce the compressed air or steam.•Versatility of employing steam or air generating equipment for other uses in aplant or process.•Relative costs of compressed air andsteam for a particular locality.•Operating requirements for the ejector,both vacuum and load. With all of thesefactors in mind, using the air-operatedejector often proves to be quite superior to other methods of producing vacuum.

HOW THEY WORK

All ejectors operate on a commonprinciple. They entrain air or other fluidsn a high velocity jet of propelling air,steam, water or other fluid. And theyuse the kinetic energy in the highvelocity stream of that fluid to push backthe atmosphere from the discharge of the ejector.

This would suggest that the higher thevelocity of the jet from the nozzle of theejector, the greater the pressure against

which the ejector can exhaust. Or if theexhaust pressure remains constant, thehigher the vacuum produced by theejector. This is true and for anyparticular velocity of the jet there is, of course, a limit to the vacuum that canbe produced.

Fig. I illustrates approximately theconversion of air pressure into velocity inthe nozzle of the ejector and theconversion of velocity into pressure inthe diffuser.

Air, under the same conditions of temperature and pressure, has less

internal energy in its molecules thansteam. And theoretically air cannotproduce as high a vacuum as cansteam. However, the inefficiencies of theexpansion and compression processesin an ejector when the ejector isoperating over its maximum range of compression obscure the differences inultimate vacuum produced.

For most practical purposes a one or two stage air ejector will produce ashigh an ultimate vacuum as will a one or

two stage steam ejector. The steam jet,however, requires fewer lbs. of motivefluid to evacuate a closed vessel thanthe air jet and fewer lb./hr. of motive fluidto exhaust a constant load at aparticular vacuum as compared to an air jet. Therefore we need to know someadditional comparative characteristics tobase our cost estimates on.

BASIS OF COMPARISON

Because 100 psig. is a very commpressure for both compressed air steam in industrial plants, it is a gpressure on which to base a comparibetween air-operated and steoperated ejectors.

200 F. is approximately the maximair temperature at which 100 psig. sinstage air compressors will deliver without requiring the compressor to excessively hot. The hotter the air toejector, the less air is required by ejector for any particular conditionvacuum and load.

If the air aftercooler of a compressobypassed or if the cooling water to aftercooler is shut off, relatively hotcan be obtained for use in an ejec

But by doing so the air storage tcapacity is reduced and condensatecollect in the storage tank and air line

This might be undesirable for socompressed air installations. It is mdesirable to heat the air by means oelectric heater or with a steam toheat exchanger. Only a very samount of electricity or low presssteam is required to reach 200 ºF.

Fig. 1 - Ejector nozzle converts air pressure into velocity and the diffuser converts

velocity back into pressure.



(or hotter), and in most cases thereduced air requirements of the ejector are well worth the additional expense.

By heating motive air to 200 F., the air required to operate an ejector can bereduced to as little as 70% of the air requirements for 70 F. air. Sometimesair ejectors are selected to keep thetemperature of the load fluid low. Thisrules out steam ejectors. And to removethe load fluid in condensers mostefficiently it would then be necessary tooperate the ejector with cold air.

TEST RESULTS

Data from our test runs on one and twostage air ejectors (of optimum design)correlate very well with data on steamejectors. We used air at 100 psig. and200 F. in our tests and compared theresults with steam ejectors operating on100 psig. dry saturated steam .

Single stage air ejectors require 1.4. -1.5 lb. of air to handle the samecondition of vacuum and load that 1.0 lb.of steam will when it is supplied to a

single stage steam ejector.

In a two stage air ejector, 2.5-2.7 lb. of air will be needed to do the same jobthat 1.0 lb. of steam will do in a twostage non-condensing steam ejector.

These ratios change somewhat whenthe pressure of the motive air ischanged. A typical figure for singlestage might be 1.7 lb. of 200 psig air per

lb. of dry saturated steam at 200 psig.Or 1.4 lb. of 60 psig. air per lb. of drysaturated steam at 60 psig .

Fig. 2 shows the ratio of motive air toload air required for one stage ejectors.The absolute pressure scale covers theoperating vacuum range of one stageunits. Fig. 3 shows the ratio of motive air to load air required for typical two stageejectors designed for any particular vacuum in the operating range for twostage ejectors. The ratios are based onsupplying motive air at 100 psig. and200 F. to remove load air at 70 F.

These ratios will be higher for load air above 70 F. and lower for load air below70 F. But the corrections are smallbetween 50-90 F. If the ejector is tohandle a fluid other than air, the flowratio must be corrected for the differencein the thermodynamic properties of theload fluid and those of air. Thiscorrection factor is usually considered afunction of the relative molecular weightsof the load fluid and air.

Fig. 2 shows that for pressure above 3.2

in. Hg abs., a single stage air-operatedejector is more economical to operatethan a two stage ejector (when themotive air pressure is 100 psig.). Theexact pressure at which two stages of compression become more economicaldepends on the pressure of the motiveair supply. Absolute pressures as lowas 0.394 in. Hg abs. (10 mm.) arepractical with a two stage air-operatedejector.

WHAT IT COSTS

Figs. 4, 5 and 6 show the operacosts of one and two stage air ejecwhen the cost of the compressed aknown.

Compressor manufacturers horganized and published much usdata which permit an analysis compressed air costs. These costs made up of: .•Operating costs including power, larepairs, maintenance, lubricants, etc•Depreciation of equipment.

•Interest on the investment made forequipment.Power is the largest portion of total c And in many cases the cost of poneed be the one consideranecessary for a study of compressedcosts.

We have used the tables ‘Compressed Air Data,” Ingersoll-RCo., Phillipsburg, N.J. (1939) compute the cost of power requiredcompressed air. The other costs, be

unique to each application, shouldstudied to determine their relaimportance and effect on the ovecost.

To use the “Compressed Air Dtables it is necessary to know the brhorsepower required to compress deliver 100-cfm. of air and the local of the various fuels under consideratio

Figs. 2 and 3 - Air consumption for single-stage and two-stage air-operated ejectors.



Since the brake horsepower will varyconsiderably with the size and type of compressor, you should obtain exact data onbrake horsepower requirements from themanufacturer after the air requirements areknown. However, typical figures are shown in atable of the reference we mentioned above. Andthe use of these figures will permit anapproximate cost analysis.

SAMPLE PROBLEM

Let’s assume that an air-operated ejector isrequired to maintain an absolute pressure of 5n. Hg in a system that has an air leakage of 25b/hr. The costs of various fuels available are:

Electricity 1.5 c./kwhFuel oil 9.5 c./galGas 63.7 c./M cu. ft.Gasoline 22.0 c./gal.

Coal $9.79/ton

Fig. 2 shows that a one stage ejector will dothe job and that 6.7 Ib of 100 psig., 200 F.motive air are required for every lb. of air to beevacuated. Therefore, the total motive air required to operate the ejector would be:

6.7lb.

motive air

X 25 lb. loadair

=

lb. load air hr.167.5 lb. motive air/hr.

We can now use Fig. 7 to find that 167.5 lb./hr.of air is equivalent to 37.5 standard cu. ft. of air per min.From our reference, the brake horsepower requirements of a typical single



stage 100 psig. air compressor with acapacity of slightly more than 37.5 scfm.s found to be approximately 22 bhp./100scfm. delivered. With this value and thefuel costs listed above we can enter theother tables of the “Compressed Air Data”book and find the power costs for runningthe compressor on the various fuels:

Electricity 0.412c. (37.5) (60)=100 cu.ft.9.27 c./hr.

Fuel Oil 0.218 c. (37.5) (60)=100 cu. ft.4.91 c./hr.

Gasoline 0.962 c. (37.5) (60)=100 cu. ft.

21.65 c./hr.Gas 0.242 c. (37.5) (60)=

100 cu. ft.5.45 c./hr.

In order to determine the cost of air compressed by a steam turbine or steamengine driven compressor, we would haveto know the steam rate of the turbineengine in lb. of steam per bhp.-hr. Atypical figure might be 28 lb. of steam per bhp.-hr. Then the power cost for theejector might be:$9.79 x 0.0733 c.-ton x (37.5) (60)

ton 100 cu.ft.$

= 16.15c./hr.

The reference table we have used isbased on evaporation rate of 7 lb. of water per lb. of coal burned. It will be necessaryto correct this for the actual evaporationrate.Our calculations show that for our assumed conditions a compressor drivenby an engine burning fuel oil would be the

cheapest way of producing the air necessary to operate the ejector (whenonly power costs are considered).

AIR COSTS ARE REASONABLE

When making cost analyses of air requirements from the reference tables,the various assumptions upon whicheach table is based should be checkedagainst the actual conditions of operation. It is likely that some particular fuel will be outstandingIy cheap due tolocal conditions. In such cases theseapproximate calculations will showconclusively which fuel is mosteconomical. Although the data above are limited toejectors operating on 100 psig., 200 F.air, we can see that power costs of air-operated ejectors can be quitereasonable.

AIR vs STEAM

Under most circumstances where steamis already available, a steam ejector would be used in preference to an air-operated ejector. Economics woulddictate the choice. If steam is notavailable, air might well be the cheaper motive fluid.There are also cases where air-operatedejectors are selected for other thaneconomic reasons. In general, air-operated ejectors are most desirable

where the heating or diluting features of the steam ejector are objectionable;where compressed air is more readilyavailable than steam; where theproperties of air are desirable as themotivating fluid.

SOME APPLICATIONS

There are many services for which an air-operated ejector is ideally suited. Pumppriming is readily done by means of anair or steam operated ejector whichoperates only long enough to exhaust the

air from the pump casing and piping. Thispermits the system to become fined withthe liquid to be pumped. The ejector isthen isolated from the system by meansof a valve. The pump is turned on. Andthe ejector air supply is turned off. Thisleaves the pump primed and ready for operation. A siphon pipe system which uses gravityto draw water or some other liquid over ahigh elevation without the use of

expensive pumps requires some inpriming to start-up. It can be primedusing an air-operated ejector operatingair from a portable or stationcompressor.The pumping of corrosive, tarry or sluliquids can be done without the usespecial pumps by means of an operated ejector.Frequently we want to recover vapor inintercondenser in its pure state, undiluand unheated. To accomplish this we use an air-operated ejector for the instage of compression to compress vapor to a pressure where it can be eacondensed. Either a steam ejector orair-operated ejector can be usedmaintain the required intercondenvacuum.

THE THERMOCOMPRESSOR

Many applications require compresair at a pressure below the availablepressure. This makes it necessarythrottle the air through an orifice or vato reduce its pressure. The costcompressing air to a high pressure then throttling to a lower pressure fparticular application can be reducedinstalling an air operated thercompressor.Working on the same principle avacuum producing ejector, thermocompressor picks up air

atmospheric pressure (or higher) andmeans of a high velocity air compresses the atmospheric air to required pressure. The saviaccomplished by the thermocompresare derived from reducing consumption of high pressure air by amount of atmospheric air that thermocompressor will entrain.Thermocompressors operating on steam and many other fluids have founwide and useful field of applicationindustry.

The rugged and simple constructionejectors along with the fact that they handle large volumes of fluids (withoutrelatively. enormous proportions of otypes of vacuum pumps) odetermines when and where an ejeshould be used. Other consideratimay, of course, outweigh the size simplicity factors. An overall picturerequirements is necessary to select best suited vacuum pump for your nee

Fig. 7 - Volume-weight conversion chart



OPTIMIZING PROCESS

VACUUM CONDENSERS

Graham Corporation

P.O. Box 719,

20 Florence Avenue

Batavia, N.Y. 14021-0719

Phone: 716-343-2216

Fax: 716-343-l 097

Email: [email protected]

Website: http://www.graham-mfg.com

INSIDE; REPRINTED FROM CHEMICAL ENGINEERING





reduced to a normal 1.25 times tube diameter

near the final tube row, which ensures that

velocities are sufficiently high to maintain

proper heat transfer.

Types of vacuum condensers

The geometries of surface condensers

generally follow three basic designs that

comply with standard nomenclatureestablished by the Tubular Exchanger

Manufacturers Assn. (TEMA; Tarrytown,

N.Y.):

1. Shellside-condensing design fixed tubesheet

type, designated as: AXL, BXM, AEL or

BEM. Figure 3 provides a clearer description

of the various “mix and match” geometries

and their designations

2. Shellside-condensing design removable

bundle type: AXS, AXU, AES or AEU

3. Tubeside-condensing design fixed tubesheet

type: AEL or BEM

Shellside condensing

Key features of vacuum condensers with

shellside condensation include:

lVapor inlet connection

lVapor distribution space above the tube

field

lMain condensing zone

l Noncondensable-gas cooling and final

condensing zone

l Noncondensable-gas outlet connection (or

vapor outlet)

lCondensate outlet connection

Condensers with shell diameters greater than

26 in. often have a longitudinal baffle that

runs virtually the entire tube length. This

type of condenser is denoted as a TEMA

crossflow “X” shell. A majority of the

condensation occurs in the tube field prior to

the longitudinal baffle.

Noncondensable gases and associated vapors

of saturation are drawn underneath the

longitudinal baffle by a low-pressure region

created by a downstream ejector, which isdesigned for that purpose. As

noncondensables and vapors are drawn

underneath the longitudinal baffle, that change

in direction separates condensate from the

vapors. Condensate drops down via gravity

to the bottom of the shell and is subsequently

drained from the unit. Meanwhile

noncondensables and associated vapors are

drawn through tubes beneath the longitudinal

baffle for additional cooling and condensation.

This separation of condensate from

noncondensables and remaining vapors

permits final cooling of

noncondensables to a

temperature below the

bulk condensate

temperature.

Furthermore, tubes

beneath a longitudinal

baffle contain the coldest

cooling water. This

enables a system designwhereby final

noncondensable gas and

the saturated vapor

outlet temperature is

below the cooling water

outlet temperature.

Units with smaller

diameter shells (less than

26 in.), denoted as

TEMA “E” shells, are

characterized by “up and

over” baffles in the final

noncondensable coolingsection. Here again, the

majority of condensation

takes place in the tube

field area before the “up and over” baffle

section. Internal geometry is such that there is

separation of the condensate from

noncondensables and vapors of saturation.

Only noncondensables and associated vapors

of saturation are drawn into the “up and

over” baffle section to ensure that heat

transfer is maximized. Once again, it is

possible to cool noncondensables to a

temperature below the cooling water outlet

temperature or below the average condensate

temperature.

In either case of shellside condensing, the

dominant design factor is to cool

noncondensables to the coldest temperature

possible, while at the same time maintaining

minimum pressure loss. Ensuring tha

noncondensables are cooled to the lowes

temperature possible minimizes the amoun

of condensable vapors that saturate thos

noncondensable gases. Effective condense

optimization requires cooling



noncondensables to within 10-15°F of the

inlet cooling-water temperature. This serves

to minimize the amount of vapors that

saturate the noncondensables and must be

handled by a downstream ejector.

Tubeside condensing

Although shellside condensation is more

prevalent, tubeside condensing may

also beused. In this case, cooling water is on the

shellside, while noncondensables and vapors

are directed through the tubes. In this

configuration, vapors and condensate remain

in intimate contact throughout the heat

transfer area and exit this area together at the

same location. The shellside is baffled (as in

any typical heat exchanger) because the

shellside fluid is simply water.

One special feature of tubeside condensers is

in the bottom head, where the condensate

drops to an outlet drain and noncondensable

gases are extracted through a connection onthe side of the head.

Noncondensable gases

Due to the sub-atmospheric condition of

vacuum systems, air inleakage is always a

potential problem. In addition, a particular

process may already have various

noncondensable gases in the process load.

With noncondensables being present,

condensation occurs along the cooling curve,

and vapors of saturation exit the condenser

along with the noncondensables.

The tube-field layout is designed to separate

condensate from noncondensables and their

vapors of saturation. It is common to have

noncondensables, along with their vapors of

saturation, exit a condenser at one location

while condensate exits another.

Flow distribution above the tube field isimportant so as to ensure that vapors and

noncondensables enter the bundle uniformly

and that there is full utilization of available

heat transfer area. Also, pressure drop is

minimized by proper flow distribution, thus

reducing utility and capital costs.

Figure 4 shows heat release curves for the

extreme cases of low noncondensable and high

noncondensable flow. Note the shape of the

respective curves and the effect that

noncondensable load has on logarithmic mean

temperature difference (LMTD), heat transfer

rate and required surface area.

Noncondensable gases serve to lower LMTD

and heat transfer rate, while consequently

increasing required surface area of the

condenser.

Precondenser pressure drop

Pressure drop in a precondenser has a

compounded impact. Depending on the

process, precondensers are positioned to

recover valued overhead vapors as condensat

prior to their introduction to an ejecto

system. As pressure drop increases, more

condensable vapors exit the precondense

with noncondensable gas. Not only does this

reduce the amount of condensable vapo

recovered, it increases the gas load to the

ejector system and its compression

requirements. As load and compression rangeincreases, so do utility requirements and

wastewater treatment costs. Pressure drop

across the intercondenser similarly increases

utility requirements for an ejector system

Table 1, p. 102, highlights the impact o

pressure drop across a precondenser.

System interdependency

Within a vacuum system, there is an

interdependency between an ejector and

intercondenser. This relationship must be

understood for optimum design and to ensur

reliable operation. An intercondenser is

designed to handle discharge load from

preceding ejector at a pressure equal to, o

below, that which is achievable by tha

ejector. Furthermore, the intercondenser



must condense the condensable vapors and cool

noncondensables in a manner that satisfies the capability of

the next following ejector.

Should an intercondenser not satisfy the dischargecapabilities of its preceding ejector or the suction capacity

of the ejector that follows it, a discontinuity occurs. The

result is that the preceding ejector ceases proper operation,

resulting in a sharp rise in the operating pressure of the

vacuum vessel, which ultimately affects product quality. It

is for this reason that ejector-condenser interdependency

must be understood and taken into account.

Equipment installationProper installation of vacuum condensers is important for

smooth operation. Typical plant layouts allow vacuumcondenser condensate to drain by gravity to a condensate

receiver. The leg height of the condensate drain must be

sufficient to ensure that condensate is not lifted into the

intercondenser because of the vacuum operation.

A straight vertical drain leg is preferred. This may not

always be possible, however. Should a layout require an

offset, horizontal runs of pipe should not be used.

Horizontal piping runs allow the formation of air pockets,

which offer additional resistance to drainage, and may

cause the flooding of a condenser.

The suggested practice is to lay out a drain leg with no lessthen a 45 deg angle, measuring from the horizontal axis,

and ensuring at least a 5ft straight length prior to the

angled run of piping. Remember to always take into

account the operating pressure of the condensate receiver.

As the condensate receiver’s operating pressure increases,

so does required drain leg height. Figure 5, above, shows

acceptable drain design.

Equipment layoutPressure drop due to piping between components is just as

important as pressure drop across a condenser. Keeping

pipe diameter equivalent to connection size on thecondenser is one key to minimizing piping loss. Also, one

should maintain interconnecting piping as short as possible

Furthermore, always try to position a precondenser or first

stage ejector as close to a vacuum vessel as possible. If a

all possible, directly connect the two items; sometimes it is

possible to mount a precondenser directly atop a vacuum

vessel. First stage ejectors may be coupled directly to the

vacuum vessel, as well.

Remember the importance and negative impact of even a

small pressure drop loss in a high vacuum processing

system. A 2 mmHg pressure loss due to piping has agreater impact on equipment size, utility and cost when tha

pressure drop is taken at 15 mm Hg absolute rather than a

80 mm Hg absolute pressure.

Edited by David J. Deutsch





Operating principleThe basic operating principle of an ejector is to convert

pressure energy of high pressure motive steam into velocity.

High velocity steam emitted from a motive nozzle is then used

to work on the suction fluid. This work occurs in the suction

chamber and diffuser inlet. The remaining velocity energy is

then turned back into pressure across the diffuser. In simple

terms, high pressure motive steam is used to increase the

pressure of a fluid that is at a pressure well below motive

steam pressure.

Thermodynamically, high velocity is achieved through adiabatic

expansion of motive steam across the converging/diverging

motive nozzle from motive pressure to suction fluid operating

pressure. The expansion of the steam across the motive nozzle

results in supersonic velocities at the nozzle exit. Typically,

velocity exiting a motive nozzle is in the range of Mach 3 to 4,

which is 3000 to 4000 ft/sec. In actuality, motive steam expands

to a pressure below the suction fluid pressure. This creates the

driving force to bring suction fluid into an ejector. High velocity

motive steam entrains and mixes with the suction fluid. The

resulting mixture is still supersonic. As this mixture passes

through the converging, throat, and diverging sections of a

diffuser, high velocity is converted back into pressure. Theconverging section of a diffuser reduces velocity as the cross-

sectional area is reduced. The diffuser throat is designed to

create a normal shock wave. A dramatic increase in pressure

occurs as flow across the shock wave goes from supersonic, to

sonic at the shock-wave, to subsonic after the shock wave. In a

diffuser diverging section, cross-sectional flow area is

increased and velocity is further reduced and converted to

pressure.

The performance curve

Ejector manufacturers summarize critical data

on a performance curve. Figure 3 shows a

performance curve for a single stage ejector.On the y-axis of this curve is suction pressure

in millimeters of mercury absolute (mm HgA).

On the x-axis is the water vapor equivalent load

(Ib/hr).

Equivalent load is used to express a process

stream, which may be made up of many

different components, such as air, water vapor

and hydrocarbons, in terms of an equivalent

amount of water vapor load. Figures 4 and 5,

from the Heat Exchange Institute Standards for

Jet Vacuum Systems, show the curves that are

used to convert various molecular weight

gases to the appropriate vapor equivalent at areference temperature of 70°F.

The performance curve can be used in two

ways. First, if the suction pressure is known for

an ejector, the equivalent vapor load it handles

may be determined. Secondly, if the loading to

an ejector is known, suction pressure can be

determined. If field measurements differ from

a performance curve, then there may be a

problem with either the process, utilities or

ejector.



Motive steamMinimum motive steam pressure is important

and is also shown on a performance curve.

The manufacturer has designed the system to

maintain stable operation with steam

pressures at or above a minimum steam

pressure. If motive steam supply pressure falls

below design, then a motive nozzle will pass

less steam. When this happens, the ejector is

not provided with sufficient energy to compress

the suction fluid to the design discharge

pressure. The same problem occurs when the

supply motive steam temperature rises above

its design value, resulting in increased specific volume,

and consequently, less steam passes through the motive

nozzle.

An ejector may operate unstably if it is not supplied with

sufficient energy to allow compression to its design

discharge pressure. Unstable ejector operation is

characterized by dramatic fluctuations in operating

pressure. If the actual motive steam pressure is below

design or its temperature above design, then, within limits,

an ejector nozzle can be rebored to a larger diameter. Thelarger nozzle diameter allows more steam to flow through

and expand across the nozzle. This increases the energy

available for compression. If motive steam supply

pressure is more than 20 - 30% above design, then too

much steam expands across the nozzle. This tends to

choke the diffuser. When this occurs, less suction load is

handled by the ejector and suction pressure tends to rise.

If an increase in suction pressure is not desired, then

ejector nozzles must be replaced with ones having smaller

throat diameters or the steam pressure corrected.

Steam quality is another important performance variable.

Wet steam may be damaging to an ejector system.

Moisture droplets in motive steam lines are accelerated to

high velocities and become very erosive. Moisture in motive

steam is noticeable when inspecting ejector nozzles.

Rapidly accelerated moisture droplets erode nozzle

internals. They etch a striated pattern on the nozzle

diverging section and may actually wear out the nozzle

mouth. Also, the inlet diffuser tapers and throat will have

signs of erosion. On larger ejectors, the exhaust elbow at

the ejector discharge can erode completely through.

Severe tube impingement in the intercondenser can also

occur but this is dependent upon ejector orientation. To

solve wet steam problems, all lines up to the ejector

should be well insulated. Also, a steam separator with a

trap should be installed immediately before an ejector

motive steam inlet connection. In some cases, a steamsuperheater may be required. Wet steam can also cause

performance problems. When water droplets pass

Maximum discharge pressureThe maximum discharge pressure (MDP), also shown on

performance curve, is the highest discharge pressure that

ejector has the ability to achieve with the given amount of mo

steam passing through the steam nozzle. If the discha

pressure exceeds the MDP, the ejector will become unsta

and break operation. When this occurs, a dramatic increas

suction pressure is common. As an example, when a sys

designed to produce 15 mm HgA pressure breaks operat

suction pressure sharply increases to 30 - 50 mm HgA. T

often causes a tower upset. Therefore, it is of paramo

importance to make sure ejectors do not exceed their MDP.

Since increasing the discharge pressure above the MDP cau

a loss of performance, it seems logical that lowering

discharge pressure below the MDP should have the oppo

affect. This, however is not the case. Ejectors with

compression ratio, discharge pressure divided by suc

pressure, higher than 2:l are called critical ejectors. Performa

of a critical ejector will not improve if its discharge pressur

reduced. This is primarily due to the presence of the shock w

in the ejector diffuser throat.

CondensersComponent partsCondensers are manufactured in three basic configuratio

fixed tubesheet, U-tube or floating head bun

Thermodynamically, these units perform identically. They d

only in ease of maintenance and capital cost. The fi

tubesheet unit, typically TEMA, AEM, BEM, AXM or BXM styles,

a bundle that is not removable from the shell. This uni

generally the least expensive to build. The major disadvantag

this type of unit is that the shellside of the condenser is

accessible for normal cleaning methods. The U-tube exchan

TEMA, AEU or BEU, is the next most economical type

construction for a removable bundle. Since the bundle

completely removable from the shell, it allows thorough cleanof the shellside as well as the tubeside. The major drawbac

the U-tube unit is that the U-bend section of the tube can mak

through an ejector nozzle, they decrease the

energy available for compression. Furthermore,

water droplets may vaporize within an ejector as

temperature increases. Vaporized water droplets

act as an additional load that the motive steam

must entrain and compress. The effect is a

decrease in load handling ability. With extremely

wet steam, the ejector may even become

unstable.



difficult cleaning of tube internal surfaces. Floating head

units, TEMA type AES, AET, AXS or AXT, are generally the

most expensive. The floating head adds complexity and

material to the return end of the condenser. These units are

advantageous because they allow complete access for

cleaning of both the shellside and the tubeside. Figure 6

indicates typical TEMA nomenclature for condenser designs.

Operating principleThe primary purpose of a condenser in an ejector system is

to reduce the amount of load that a downstream ejector must handle. This greatly improves the efficiency of the

entire system. Often condensers are analyzed like shell and

tube heat exchangers which are common throughout

refineries. Although vacuum condensers are constructed

like these exchangers, their internal design differs

significantly due to the presence of two phase flow and

vacuum operation.

Vacuum condensers for crude tower applications generally

have the cooling water running through the tubes. The

condensing of the water vapor and hydrocarbons takes

place on the shellside. Generally, the inlet stream enters

through the top of the condenser. Once the inlet stream

enters the shell, it spreads out along the shell and

penetrates the tube bundle. A major portion of the

condensibles contained in the inlet stream will change

phase from vapor to liquid. The liquid falls by

gravity and runs out of the bottom of the

condenser and down the tail leg. The

remainder of the condensibles and the

noncondensibles are then collected and

removed from the condenser through the

vapor outlet.

Vapor is removed from the condenser in two

ways. In larger units, approximately 30 in. in

diameter and larger, a long air baffle is

used. The long air baffle runs virtually the full

length of the shell and is sealed to the shellto prevent bypassing of the inlet stream

directly to the vapor outlet (Figure 7). This

forces the vapors to go through the entire

bundle before they can exit at the vapor

outlet.

Similarly, smaller units use an up and over

baffle arrangement to maximize vapor

distribution in the bundle. In this

configuration, the exiting vapor leaves the

condenser on one end only. The vapors are

forced through a series of baffles in order to

reach the vapor outlet. Figure 8 illustrates a

typical AEM cross-sectional drawing.

Both the long air baffle and the up and over baffles are normally located in the coldest

cooling water pass in order to guarantee

counter current flow, and cooling of vapors

and noncondensibles below exiting water

temperature and optimal heat transfer.

As mentioned previously, a condenser is

designed to limit the load to the downstream

ejector. In many cases, the load to a

condenser is ten times the load to the

ejector. Consequently any loss in condenser

performance will have a dramatic affect on





Proprietary design procedures incorporate the following

considerations:

• Condenser vapor inlet location and distribution area

above the tube field so as to insure proper vapor entry to

the shell and penetration into the tube field.

• Tube field layout and penetration areas to guarantee that

flow distribution into the bundle is well maintained and

pressure drop is held to a minimum.

• Noncondensible gas cooling section, where bulk

condensate is separated from the vapor and finalcooling to design saturation temperature is achieved.

• Bulk condensate and noncondensibles exit the shell at

different locations and temperatures. In this way,

noncondensibles and vapors are cooled below the

condensate temperature to maximize condensation

efficiency without contending with excessive condensate

loading and associated thermal duty.

• Support plate spacing and bundle penetration areas to

insure velocities are well below those necessary to

establish vibration.

• Process vapors assessed to properly ascertain

vapor/liquid equilibrium (VLE) conditions throughout the

condensing regime.• Condensing profile broken down into as many as fifty

steps to properly determine the effective LMTD and VLE

at each step.

Often proprietary designs are compared to those

determined by computer programs available from

institutional organizations, research companies or

software companies. These generic programs do not

properly model flow configurations typical of vacuum

condensers. A number of organizations put forth excellent

software to reliably predict performance of process heat

transfer equipment, however, that same software should

not be applied to exchangers designed for vacuum

condensation. The software is unable to model internalconfigurations typical of vacuum condensers and they

typically force condensate and noncondensibles to exit the

same connection and be at the same temperature.

The ejector system

Type of tower As mentioned above, typical operating modes for a vacuum

tower are classified as wet, damp or dry.

Wet towers have overhead loading characterized by

substantial amounts of stripping steam plus typical

amounts of coil steam to the fired heater. Operating

pressure for a wet tower has a range of 50 - 65 mm Hg

Abs at the tower top and a flash zone pressure of

approximately 65 - 75 mm Hg Abs. With such moderate

vacuum levels, often it is possible to have a precondenser

between the vacuum tower and a two stage ejector system.

The precondenser reduces loading to the ejector system

by condensing substantial amounts of steam and

hydrocarbon vapors, thereby reducing energy demands to

operate the ejector system.

• A damp tower operates typically in the range of 15-25

mm Hg Abs at the tower top, with flash zone pressure of

approximately 35 mm Hg Abs. Stripping steam is

appreciably reduced and the ejector system is a three

stage system.

• Dry towers operate between 5-l5 mm Hg Abs at the

tower top, flash zone pressure at 20 mm Hg Abs, and do

not utilize stripping steam. Here again, it is customary to

utilize 3 stage ejectors. It is not possible to operate at

these pressures and utilize a precondenser. Theoperating pressure is below a level where cooling water

is cold enough to induce condensation. There are cases

of deep-cut operation where the pressure may be below

5 mm Hg Abs and a 4 stage ejector system is used.

Here two ejector stages are in series ahead of the first

intercondenser (Figure 9).

Ejectors/condensersFrom the figures referenced above, it is understood that

ejectors and condensers are staged in series with each

other. Process vapors and noncondensibles flow in series

from the tower to an ejector, then to an intercondenser,

followed by another ejector, then to an intercondenser, etc.The purpose of an ejector is to entrain tower overhead

vapors and noncondensibles, and then compress them to

a higher pressure. Ultimately, via a series of staged

ejectors, process fluids are brought to a pressure

equivalent to atmospheric pressure or greater. For

example, a vacuum tower is maintained at 10 mm Hg:

• 1st stage ejector compresses process fluid from 10 - 80

mm Hg.

• 2nd stage ejector compresses from 80 - 250 mm Hg.

• 3rd stage ejector compresses from 250 - 800 mm Hg.

The purpose of intercondensers, as mentioned previously,

is to be positioned between ejector stages to condense as

much steam and hydrocarbons as possible. By

condensing steam and hydrocarbon vapors, the load

handled by a downstream ejector is reduced. This

maintains energy usage (motive steam consumption) for

driving the ejectors, to a minimum.



Process conditionsThese are very important for reliable vacuum system

operation. Process conditions used in the design stage

are rarely experienced during operation. Vacuum system

performance may be affected by the following process

variables, which may act independently or concurrently:

• Noncondensible loading. Vacuum systems are

susceptible to poor performance when noncondensible

loading increases above design. Noncondensible

loading to a vacuum system consists of air leaking intothe system, lightened hydrocarbons, and cracked gases

from the fired heater. The impact of higher than design

noncondensible loading is severe. As non-condensing

loading increases, the amount of saturated vapors

discharging from the condenser increases. The ejector

following a condenser may not handle increased

loading at the condenser design operating pressure.

The ejector before the condenser is not designed for a

higher discharge pressure. This discontinuity in

pressure causes the first ejector to break operation.

When this occurs, the system will operate unstably and

tower pressure may rapidly rise above design values.

• Noncondensible loadings must be accurately stated. If

not, any vacuum system will suffer performance

shortcomings. If noncondensible loadings are

consistently above design, then new ejectors are

required. New condensers may be required depending

on severity.

• Condensible hydrocarbons. Tower overhead loading

consists of steam, condensible hydrocarbons and

noncondensibles. As different crude oils are processed

or refinery operations change, the composition and

amount of condensible hydrocarbons handled by the

vacuum system vary. A situation may occur where the

condensible hydrocarbon loadings are so different from

design that condenser or ejector performance is

adversely affected. This may occur in a couple of different ways. If the condensing profile is such that

condensible hydrocarbons are not condensed as they

were designed to, then the amount of vapor leaving the

condenser increases. Ejectors may not tolerate this

situation, resulting in unstable operation. Another

possible effect of increased condensible hydrocarbon

loading is an increased oil film on the tubes. This

reduces the heat transfer coefficient. Again, it may result

in increased vapor and gas discharge from the

condenser. Unstable operation of the entire system may

also result. To remedy performance shortcomings, new

condensers or ejectors may be necessary.

Tower overhead loading. In general, a vacuum system will

track tower overhead loading as long as noncondensibleloading does not increase above design. Tower top

pressure follows the performance curve of the first-stage

ejector. Figure 3 shows a typical performance curve. At light

tower overhead loads, the vacuum system will pull tower

top operating pressure down below design. This may

adversely affect tower operating dynamics and pressure

control may be necessary. Tower pressure control is

possible with multiple element trains. At reduced overhead

loading, one or more parallel elements may be shut off.

This reduces handling capacity, permitting tower pressure

to rise to a satisfactory level. If multiple trains are not used,

recycle control is another possible solution. Here, the

discharge of an ejector is recycled to the system suction.

This acts as an artificial load, driving the suction

pressure up. With a multiple-stage ejector system,

recycle control should be configured to recycle the load

from before the first intercondenser back to system

suction (Figure 10). This way, noncondensible loading

is not allowed to accumulate and negatively impact

downstream ejectors.

• System back pressure . Vacuum system back pressure

may have an overwhelming influence on unsatisfactory

performance. Ejectors are designed to compress to a

design discharge pressure (MDP). If the actual

discharge pressure rises above design, the ejectors will

not have enough energy to reach the higher pressure.

When this occurs, the ejector breaks operation and

there is a sharp increase in suction pressure. When

back pressure is above design, possible corrective

actions are to lower the system back pressure, rebore

the steam nozzle to permit the use of more motive

steam or install a completely new ejector.

InstallationSufficient clearance should be provided to permit removal

of the motive chest which contains the motive nozzle whichprotrudes into the suction chamber. The ejector may be

installed in any desired position. If the ejector is pointed

vertically upward, a drain must be present in the motive

chest or in the suction piping to drain any accumulated

liquid. This liquid will act as load until it is flashed off,

giving a false performance indication. The liquid could also

freeze and cause damage. The motive line size should

correspond to the motive inlet size. Oversized lines will

reduce the motive velocity and cause condensation.

Undersized lines will result in excessive line pressure drop

and, thus, potential low pressure motive to nozzle. The

motive fluid lines should be insulated.

The suction and discharge piping should match or belarger than that of the equipment. A smaller size pipe will

result in pressure drop possibly causing a malfunction or

reduction in performance. A larger pipe size may be

required depending on the length of run and fittings

present. Appropriate line loss calculations should be

checked. The piping should be designed so that there are

no loads (forces and moments) present that may cause

damage. Flexible connections or expansion joints should

be used if there is any doubt in the load transmitted to the

suction and discharge flanges. If the system vent is

designed to exhaust to a hotwell, the pipe should be

submerged to a maximum of 12 in. If the discharge



exhausts to atmosphere, the sound pressure level should

be checked for meeting OSHA standards, paragraph 1910.95

and Table G-12 and/or the local standards.

A thermostatic type condensate trap should be avoided since

they have a tendency to cause a surge or loss of steam

pressure when they initially open. This could cause the

ejector to become unstable.

Operation

Start-upThe ejector motive line should be disconnected as near as

possible to the motive inlet and the lines blown clear. This is

extremely important on new installations where weld slag

and chips may be present and scale particles could exist.

These particles could easily plug the motive nozzle throats. If

a strainer, separator, and/or trap is present they should be

inspected and cleaned after the lines are blown clear. The

vapor outlet of the aftercondenser and condensate outlets

should be open and free of obstructions and the cooling

medium should be flowing to the condenser(s).

All suction and discharge isolat ing valves, if present, should

be opened. If the unit has dual elements with condensers

present, ensure the condenser is designed for both

elements operating. If the condenser has been designed for

one element operating, the suction and discharge valves

should be opened to only one element (the other element

being isolated).

The motive valve to the last ejector stage (‘Z’ stage) should

then be fully opened. For optimum performance during an

evacuation cycle the motive valves should always be opened

starting with the ‘Z’ stage and proceeding to the ‘Y’, ‘X’, etc.

stages. If a pressure gauge is present near the motive inlet,

the reading should be taken to ensure the operating

pressure is at or slightly above that for which the unit is

designed. The motive pressure gauge should be protected

with a pigtail to insure protection of the internal working parts

of the gauge. The design operating pressure is stamped onthe ejector nameplate.

ShutdownThere are two procedures to be considered when shutting

down: method A is appropriate if it is desired to maintain the

vacuum upstream of the first stage ejector (an isolating valve

has to be present at suction) rather than allow pressure to

rise to atmospheric pressure, in which case the valves

should be closed in the following order:

• Close 1st stage suction valve.

• Close 1st stage motive inlet valve.

• Close 2nd stage suction valve.

• Close 1st stage discharge valve.

•Close second stage motive inlet valve.

• Close 2nd stage discharge valve (if present).

If there are more than two stages, then the second stage

motive inlet valve should be closed on all ejectors before the

second stage discharge valve is closed. If the system

contains an isolating valve at the first stage suction only, the

procedure would be to close this valve and then either shut

off the motive to all ejectors at once or shut them off by stages

starting at the first stage. When all the motive valves have

been shut off, the cooling medium may be turned off. If the

unit is going to be shut down for a short period of time to

service the ejectors or for some other reason, it is not

necessary to shut off the cooling medium. Energy savings

should be considered when making this decision. If the unit is

going to be down and freezing of the cooling medium is

possible, then measures must be taken to prevent freezing or

the unit drained as much as possible to prevent damage.

Allowing a small amount of coolant to continuously flow will

usually prevent freezing.

Method B is employed if it is not required to maintain a vacuum

upstream of the first stage ejector and the valves should be

closed in the following order:

• Close motive valve to all ejectors or close the motivevalve(s) to each individual stage starting at first stage and

continue on to second, etc.

• The cooling medium may be turned off as explained in the

preceding paragraphs.

Switching ejector elementsShould it become necessary or desirable to shift from one two

stage element to another while the unit is in operation, then

the procedure is as follows:

• The standby Z stage ejector discharge valve (if provided)

should be opened.

• The Z stage motive valve should then be opened.

•The Z stage suction valve should then be opened. Whenthis has been accomplished, this standby Z stage ejector

begins to take suction from the intercondenser along with

the other Z stage element.

• The Y stage discharge valve on the standby element should

then be opened.

• This is to be followed by opening the Y stage motive valve.

• The Y stage suction valve should then be opened. At this

point both two-stage elements are in parallel operation. The

procedure then continues as normal. The operating

element can now be secured by closing the valves as

follows:

• Close 1st stage suction valve.

• Close 1st stage motive valve.

• Close 2nd stage suction valve.• Close 1st stage discharge valve.

• Close 2nd stage motive valve.



• Close 2nd stage discharge valve (if

provided).

Again the sequence then continues as

normal.

Operating surveyThe goal here is to introduce a systematic

way to troubleshoot a crude vacuum system.

The first task is to review design data and

then go out into the field and take data. This

leads to the most important part of vacuum

system troubleshooting: how and what data

should be taken.

Figure 11 shows the appropriate test points

for a three stage crude vacuum system. The

following test points are mandatory for proper

system troubleshooting:

• Suction and discharge pressure on each

ejector.

• Motive steam pressure at each ejector.

• Cooling water inlet and outlet pressures for

all condensers.

• Cooling water inlet and outlet temperaturesfor all condensers.

It is essential that all of these readings are

accurate. The most common cause of

misdiagnosing vacuum system problems is

inaccurate or inconsistent measurements.

For this reason, certain guidelines must be followed.

Accurate suction and discharge pressures at each

ejector are the most important and most difficult

readings to take.

All ejector suction and discharge pressures, except for

the last stage discharge pressure, will be in the range

from I - 400 mm HgA. Measuring pressure in this range

requires a high accuracy absolute pressure gauge.

Wallace & Tiernan absolute pressure gauges arecommonly used. This gauge should not be permanently

mounted to the system. It should be kept in a lab until it

is needed. All absolute pressure measurement devices

are delicate and prone to being knocked out of

calibration by process vapors and liquids. A common

compound pressure gauge with a range of 30 in.

HgV/0/30 psig is often used by refinery personnel to take

these measurements. This type of gauge is simply not

accurate enough to yield useful vacuum measurements.

The motive steam pressure and cooling water inlet and

outlet pressures should be measured with a properly

ranged and calibrated pressure gauge. The cooling

water temperatures should be taken with a bi-metallicthermometer using thermowells. All of the vacuum,

motive, steam, cooling water pressure and temperature

measurements should be taken with one instrument.

For instance, the steam pressure measurement should

be taken at the first stage ejector. The same gauge

should then be physically moved to the second stage

ejector and then to the third stage ejector. This

eliminates any possible difference in gauges caused by

wear, over pressurization, shock, etc. Quite often, small

ball valves are permanently added to the equipment to

facilitate this type of testing.



Table 2 is a compilation of design and

test data taken for the three stage

crude system shown in Figure 11. The

column marked ‘Design’ shows the

design values for all the test points.

The design suction, discharge and

motive pressures, P1-9, are all taken

from the system performance curve

shown in Figure 12. The ejector

discharge pressures are calculated

from the curve assuming a maximumpressure drop of approximately 5%

across each condenser. The design

values for condenser inlet and outlet

cooling water temperature and cooling

water pressure drop, ∆p, are obtained

from the manufacturer’s condenser

data sheets. As shown, there are no

design values given for the cooling

water inlet and outlet pressures. For

design and troubleshooting the only

Measurement data can then be compared to the design

data. This is done using the system performance curve

and data sheets. It is often very helpful to be able tocompare new data to baseline data taken when the system

was operating correctly

important number is the pressure loss across the

condenser, not the actual pressure.

Case studies 1 to 4 represent examples of different types



of common performance problems. In each case, a

different problem was found with the equipment. After

each case has been dicussed, there will be an additional

section on how mechanical failures can also contribute to

the symptons shown.

Case study 1:

fouled condenser The most common performance problem with steam

ejector systems is lower than design steam pressure. For

this reason, motive steam pressure is always the first data



steam pressure is always the first data that should be

examined. In this case, the motive steam pressures at

each ejector, P7-9, are all above design and should not

pose any performance problems. Next, the ejector suction

and discharge pressures are examined, starting with the

third stage ejector. The process begins with the last stage

because if that is not working, then the other stages will not

work either.

Here, the third stage discharge pressure, P6, and third

stage suction pressure, P5, are both below design. Thus,

the third stage ejector is operating correctly and its loadmust be within design limits. Since the third stage ejector

load is within design limits, the second intercondenser

must be working properly. Next, the second stage ejector

discharge pressure, P4, is examined. It is also below

design, indicating an acceptable shellside ∆P of 3.5%.

Remember, pressure drop across a vacuum condenser

should be less than 5% of its operating pressure.

Moving to the second stage ejector suction, P3, the

system’s problems begin to show up. P3 is 13 mm Hg

higher than design. It is not possible for the first stage

ejector to compress its load to 96 mm Hg Abs, 13 mm Hg

greater than the 83 mm Hg Abs design, and still maintain a

suction pressure of 20 mm Hg Abs. The higher thandesign first stage discharge pressure is causing the first

stage ejector to break operation. The logical cause of the

high second stage ejector suction pressure is a fouled first

intercondenser. To confirm this, the cooling water data is

examined.

The cooling water pressure drop on all three condensers

is normal, indicating cooling water flow rate is

approximately at design. The cooling water temperature

rise is low across the first intercondenser and high across

the second intercondenser. The low temperature change

on the first intercondenser indicates that the cooling water

is not absorbing as much heat as it should and therefore,

must be fouled. As previously discussed, a fouled

condenser allows greater vapor carry over to thedownstream ejector. This accounts for the high second

stage ejector suction pressure and high second

intercondenser cooling water temperature rise.

Case study 2:excessive noncondensible loadingFollowing the same thought process as case study 1,

motive steam pressure is not a problem. The third stage

ejector discharge pressure is also under design. It is

noted that the third stage ejector suction pressure is higher

than design, measured at 305 mm Hg Abs versus a

design of 277 mm Hg Abs. This appears to affect first and

second stage ejector performance.

Possible causes of an elevated suction pressure arecooling water flow rate below design, cooling water inlet or

outlet temperature greater than design, condenser fouling

or higher than design loading to the ejector. Reviewing

cooling water data suggests no abnormalities, i.e.

pressure drop across each condenser seems acceptable

and cooling water temperatures are below design values.

With cooling water pressure drop and temperature rise at

each condenser close to design values, fouling may be

ruled out. The remaining possible cause is an increased

load to the ejector.

Common performance problems arise when

noncondensible gas loading exceeds the design value.

Higher non-condensible loading results in increased

loading to downstream ejectors. This is due to a higher

mass flow rate of noncondensibles plus their associated

vapors of saturation.

The elevated pressure at the third stage ejector suction

causes the second stage to break operation. Again, this is

because the second stage ejector is unable to compress

its load to a pressure greater than 292 mm Hg Abs.

Therefore, there is an increase in the suction pressure of

the second stage as it breaks operation. This, in turn,

forces the first stage to break operation and the suctionpressure to the system increases from 20 mm Hg Abs to

62 mm Hg Abs.

Case study 3:excessive condensible loading

This case is characterized by a modest loss in lower top

pressure. Once again, the steam pressure to each ejector

is satisfactorily above design. The third stage ejector

suction and discharge pressures are below design. The

second stage ejector suction and discharge pressures are

also below normal, as is the first stage ejector discharge

pressure. The only pressure that is abnormal is the first

stage ejector suction pressure.

The cooling water data indicates all three condensers havehigher than design cooling water pressure drops and

lower than design temperature rises. This indicates that:

the high cooling water pressure drop is an indication of

either fouling or high cooling water flow rate. The low ∆T

indicates that either the condensers are fouled or that there

is a high cooling water flow rate. The previous analysis of

the suction pressures of the second and third stage

ejectors show no signs of fouling, i.e. elevated suction

pressures. The conclusion must be that there is a higher

than design cooling water flow rate to the condensers.

Higher cooling water flowrate does not affect ejector

system performance. The elevated first stage suction

pressure and tower top pressure must be the result of a

high condensible load causing the ejector to run out further

out on its curve.

Case study 4:low motive steam pressureUsing the same method as previous case studies

provides a quick answer to this performance problem. The

steam pressure on the second stage ejector is below

design. As discussed earlier, this will cause the second

stage to break operation. When this second stage ejector

breaks operation, its suction pressure rises above the

maximum discharge pressure of the first stage ejector.

This results in broken operation for the first stage ejector

and increased tower top pressure. This situation will

correct itself if the second stage ejector steam pressure isincreased.

Mechanical problemsNow that examples of how process conditions, fouling and

utilities will affect system performance have been seen, it

needs to be understood what affect mechanical problems

will have on a system. A common mechanical problem is a

loose steam nozzle. When a steam nozzle becomes loose

it begins to leak steam across the threads. The leaking

steam then becomes load to the ejector. If the loose nozzle

occurs in the first stage ejector the affect will be an

overloaded first stage ejector. If the leak occurs in the





Proper Piping for Vacuum Systems

LOREN WETZEL

GRAHAM MANUFACTURING CO.

Optimally designed piping upstream and downstream of vacu-um equipment increases equipment efficiency and reduces

maintenance. It also minimizes vacuum loss and pressure drop, takesadvantage of suction lift to enhance energy efficiency and decreasesthe risks of flooding equipment or shutting down systems.

Unfortunately, however, contractors or engineering firms doingplant layout frequently either route piping to accommodate exist-ing process equipment, or try to fit pipes into available space.

Such slipshod piping configuration contributes greatly to plantdowntime and process inefficiency.

In addition, many plant startups and modifications are delayedbecause a simple piping installation had been performed improp-erly. And, if a problem is found after startup, it may not berectifiable without considerable trouble and expense. This articlediscusses the principles of proper piping design for common plantequipment, such as tailpipes, hotwells and float traps.

Trapped bubbles in tailpipes. A common hazard in barometric orshell-and-tube condenser tailpipes is accumulating gases.Condensate from a shell-and-tube condenser, or cooling waterplus condensed steam or hydrocarbons from adirect-contact barometric condenser, always con-tain air or other non-condensible gases.

A horizontal or slightly downward-sloped line isvulnerable to these gases, which cling to upper pipesurfaces. All types of pipe contain a certain amountof internal roughness and, because of this, gasestend to start clinging and building up in the small-est crevice. In addition, every flanged joint has aslight crack where a gasket is located, thus permit-ting another place for gases to collect.

As these gases accumulate, they form tiny bubbles,growing into larger ones that eventually becomebig enough to partially or completely block off piping at that point. The condensate cannot flowdownwards and soon its level rises, flooding thecondenser.

Testing has proven that if piping changes direc-tion, it must form at least a 45-deg angle from thehorizontal (Figure 1). With this amount of sloping,gases will either slide back up the pipe or continue

downward with the thrust of the flow-ing water. Observe that this is truewhether the condenser is a barometricor shell-and-tube unit.

When a change in direction is required,there must always be a vertical straightdistance of five pipe diameters or four ftminimum between each change. This

allows flowing liquid to develop a mini-mum velocity head and a straightdownward pattern before the first change in direction. There areno valves in the tailpipes shown (Figure 1), for two reasons:

• If a valve is accidentally left closed during startup or on turn-around, or if vibration closes a valve partly or completely, thecondition can flood condensers, cause vacuum loss and shutdown operation

Chemical Engineering, November 1996 1

Figure 1 (top). If piping must change directi on, i t should form at least a 45-deg angle

from the horizontal plane; the horizontal piping in the ri ghtmost drawing is vulnerable

to gas accumulati on.



• Any valve, by definit ion, causes pressure drop. Unlike asmooth piece of pipe, a valve creates a node, in which prod-

ucts such as hydrocarbons, salts or rust can accumulate. Thisleads to excessive pressure drop, or can result in closing off piping completely and possibly shutting down operations

CONFIGURING FOR SUCTION LIFT

Suction lift is a function of vacuum systems that can be used toadvantage in piping (Figure 2). For example, it can enhance apumping system by reducing the load on an existing motor.

Imagine, for instance, pumping a liquid from one level up 80 ftto a vessel operating under vacuum. The vacuum or suction liftcan be used to reduce the total dynamic head (TDH) require-ments for the system’s pump and motor.

This reduces the horsepower used and possibly the motor size,thus saving energy and money. Another application is to merelymove liquid from one tank to another without a pump.

To find a specific value for a given piece of equipment with Figure 2,use the lowest expected condenser pressure at the minimum cooling-water temperature at the inlet (for barometric systems), or theminimum condensing pressure due to loading. The barometric pres-sure, in addition to the absolute pressure in the condenser, greatlyaffects the suction lift. I recommend using the highest recorded baro-

metric pressure for calculation, and taking 80% of the theoreticalsuction lift to cover any overlooked condition.

For an actual check of suction lift, obtain the barometric pressuredirectly at the installation point, and measure the condenser orvessel absolute pressure. Using Figure 2, move vertically upwardfrom the actual condenser pressure reading, to the barometricpressure. At the intersection, move horizontally to the left to readsuction lift in ft H20.

TAILPIPE HEIGHTS

Recommended minimum effective tailpipe heights are shown,based on water at 32°F (Table, opposite page). This height shouldbe based on the absolute maximum recorded barometric pressurefor given equipment, regardless of the anticipated condenser oper-ating pressure. This pressure information must be used in pipingdesign when vacuum equipment is placed in a building or an ele-vated structure.

For example, consider an installation site with a highest recordedbarometric pressure of 30 in. Hg. The plant has been laid out,and the most-economical placement of the vacuum vessel (assumea process precondenser) is at an elevation of 32 ft, next to theevaporator. Based on the 30-in.-Hg maximum pressure, the mini-mum effective tailpipe for water should be 34 ft.

The result, however, is that water will flood the pre-condenser by2 ft. As something must be changed, the logical solution is tomove the evaporator and condenser to the next floor level, or toelevate them enough to overcome the difference.

Note that the values in this chart are based on water; heightsshould be corrected if any hydrocarbons or other substances arepresent. For hydrocarbons, good installation practice is to use atleast 45 ft, regardless of barometric pressure.

It is difficult to predict actual heights needed for hydrocarbonsunder vacuum. Some have a tendency to foam, which suggeststhe rule-of-thumb minimum of 45 feet. If the specific gravity of the liquid in the tailpipe is known, the height should be adjustedaccordingly.

HOTWELL DESIGN

The designer mustcarefully consider openhotwell design in aprocess (Figure 3).Good practice recom-mends that the hotwellarea be equal to 1.5times the tailpipe volumemeasured from the bot-tom of the tailpipe to thepoint of overflow (not less than 12 in.). The large volume is need-

ed to ensure there is enough liquid present to seal the tailpipe.

As vacuum is produced, the water rises in the tailpipe to theheight induced by the vacuum, minus the barometric pressure. If there is insufficient hotwell area present, the seal will be brokenand air drawn into the tailpipe, affecting the performance of vac-uum-producing equipment and the process. The pressure couldrise dramatically, affecting the process pressure, and possibly shut-ting down plant operations.


Figure 2. Use the absolute pressure of a condenser, plus baromet-

ric pressure, to estimate sucti on-l i ft values.

Figure 3. Suffi cient hotwell area is neces-

sary to contain vacuum in a tai lpipe.



LOOP SEALS AND FLOAT TRAPS

Using an intercondenser to remove condensate from an ejector toanother condenser operating at a lower pressure is a typical piping

configuration that can frequently be problematic. However, fol-lowing a few simple guidelines will eliminate problems. Theconfiguration discussed in the following paragraphs should beused primarily for turbine-exhaust condensers and their associatedinter- and inter-after condensers.

Whenever hydrocarbons are present that will condense in theinter- or inter-after condensers, or when the vacuum system is ona platform elevated about 40 ft in the air, a condensate receiver orseal tank should be used (leftmost diagram of Figure 1).

If a float trap is used (Figure 4), the intercondenser should be atleast 18 in. above the normal liquid level of the condenser intowhich condensate is dumped. I f a loop seal is used, the loop-sealheight should be equal to the difference between the highestoperating pressure in the intercondenser minus the main con-denser’s lowest operating pressure.

In looking at the highest intercondenser pressure, the designershould also consider off-design or startup conditions. In addition,the designer should take into account extremely small loads to themain condenser when using the coldest condensing-water temper-ature. This will yield the lowest main-condenser pressure.

Since piping is relatively inexpensive, loop-seal height should notbe shortened to save a few dollars. Generally, an 8- to 10-ft loopseal should be adequate; but this height should be determined bythe manufacturer of the ejector or condenser. The valve at thebottom of the tailpipe is for draining the unit when it is idle, toprevent freezing or rusting, and to service the tailpipe equipment.

Frequently, the designer runs into a space problem, requiring thatthe ejector condenser be located below the normal liquid level inthe hotwell of the condenser. This could be a problem if piping isconfigured as in Figure 5 — condensate will not flow out of theintercondenser because there is insufficient piping distancebetween the two condensers to allow this. The inter- or inter-aftercondenser will be flooded on the shell side losing vacuum andshutting down the system.


Figure 4 (top, left). Don’t skimp on loop-seal height in order to

cut costs.

Figure 5 (top, ri ght). In this incorrect example, the inter- or

inter-after condenser wi ll be flooded on the shell side.

Figure 6 (left ). The addition of a steam-powered pump corrects

the defi ciencies of Figure 5.



Such a problem can beresolved (Figure 6).Basically, this configu-ration requires apressure-poweredpump, which runs onsteam. The pump sizeand steam pressureand quantity required

are functions of totallift and actual lb/h of condensate to bepumped.

Depending on thesteam pressure avail-able, lift can be as highas 300 ft—though theneeded height is typi-cally only 8 to 15 ft,requiring relatively

low-pressure steam of 50 psig or less. The designer should always

try to pipe equipment relatively simply, as shown in Figure 4,because additional hardware (such as a pressure-powered pump)may be needed, adding to the complexity of existing piping.

Two other equipment configurations are useful when space is at apremium. First, a barometric configuration has its shell bodyextended to form a storage tank, with a level controller modulat-ing an overboard valve, plus a condensate pump removing liquidin the storage area (Figure 7). This setup is often called a “low-level barometric.”

An off-shoot of this is shown (Figure 8) with the same storageand controls, but with a shell-and-tube intercondenser mountedon top. The condensate pump, in both cases, must be carefullysized for the net positive suction head (NPSH)available.

Both of these examples are extensively usedthroughout industry. The designer, as stated, mustcarefully look at the pump NPSH, but generally asuction head of 4-5 ft is adequate. The only otherdesign criterion is sizing the control valve to satisfydownstream conditions.

HYBRID SYSTEMS

Some designs feature ejectors with a shell-and-tubeintercondenser plus a liquid ring vacuum pump(LRVP). In such configurations, the LRVP mustbe located directly below the condenser (Figure 9).

This system, commonly called a “hybrid system,”is very cost effective. As the LRVP is locateddirectly below the condenser, this application elim-

inates a second shell-and-tube intercondenser, possibly a shell andtube aftercondenser, and two additional steam-jet ejectors, realiz-ing considerable space savings.

Note, however, that an LRVP is limited, because it is pumpingcondensate as well as any noncondensible gases. An LRVP canonly pump a percentage of condensate, compared to the seal liq-uid required. Eachindividual system

should be analyzed forits particular limita-tions.

Note, also, that a sin-gle-stage ejector, or asmany as four stagesupstream of the inter-condenser, could berequired in somecases. Figure 9 uses atwo-stage configura-tion simply to depict

the principle of thesystem.

PROTECTINGAGAINSTCONDENSATE

Vapor piping entering and leaving condensers in a vacuum systemwith condensibles present can result in serious operating problems if designed incorrectly (Figure 10). With barometric condensers (Figure10a), it is important to note that condensate is splashing down thebarometric walls and could run down the vapor inlet, unless the inletis protected by a dam or series of elbows.


Figure 9. A li quid-r ing vacuum pump eliminates a second shell -and-tube intercon-

denser, as well as steam-jet ejectors.

Figure 7. A ‘low-level barometr ic” con-

fi gurati on has an extended shell body to

form a storage tank.

Figure 8. A vari ation on the low-level

barometr ic, the low-level shell -and-tube

configurati on adds an intercondenser on top



If the process vessel is a turbine, liquid can run down the pipe fromthe barometric condenser, tearing apart turbine blades, causing seriousdamage and major expense plus a shutdown. Even with a less-criticaltype of process vessel, such as an evaporator, water can contaminateproduct, increase process load or ruin product completely.

Condensible vapors flowing in a pipeline will naturally condensesince the pipe is usually cooler than the saturation temperature of the vapor it contains. Vapor piping entering and leaving a baro-

metric condenser (or a shell-and-tube condenser) must notcontain any pockets where this liquid can accumulate. This liquidwill add another flashed load to the ejector, or could seal off theline completely, resulting in a downgraded system.

The absolute pressure upstream of a pocket will rise dramatically,indicating that ejectors are not working satisfactorily. This will causea false alarm, while equipment may actually be performing properly.

Edited by Irene Kim

AUTHOR

Loren E. Wetzel is assistant manager of contract engineering forGraham Manufacturing Company, Batavia, NY. He received hisB.S. degree in mechanical engineering from Rochester Institute of Technology in 1956. He has been employed full time at GrahamManufacturing Inc. since graduation. During initial employment,he was trained in every department in the fabrication shop, aswell as both the heat exchanger and vacuum engineering depart-

ments. He specialized in ejector design. including being in chargeof the ejector testing and service departments. During this time,he was also a senior contract engineer. He has been involved atlength in the research and development of ejectors.


Figure 10. Vapor inlet piping should prevent condensate from

splashing down barometr ic walls (a); inlet and outlet piping should

not have any pockets in which condensed liquid can accumulate (b)



Understand vacuum-system fundamentalsProperly operating ejectors and condensers is importantin maximizing vacuum tower gas-oil yield

G. R. MARTIN, PROCESS CONSULTING SERVICES, Grapevine, Texas J. R. LINES, GRAHAM MANUFACTURING CO., INC.,Batavia, New York

S. W. GOLDEN,Gli tsch, Inc., Dallas, Texas

Crude vacuum unit heavy vacuum gas-oil (HVGO) yield issignificantly impacted by ejector-system performance, espe-

cially at conditions below 20 mmHg absolute pressure. A deepcutvacuum unit, to reliably meet the yields, calls for proper design of all the major pieces of equipment. Understanding vacuum ejectorsystem impacts, plus minimizing their negative effects equalsmaximum gas yield. Ejector-system performance may be adverselyaffected by poor upstream process operations.

The impacts of optimum ejector performance are more pro-nounced at low flash-zone pressures. Gas-oil yield improvementsfor small incremental pressure reductions are higher at 8 mmHgthan at 16 mmHg. Commercial operation of a column with a 4.0mmHg top pressure and 10 mmHg flash-zone pressure is possi-ble. Designing a deepcut vacuum unit calls for a balance betweenpractical limits of furnace design, column diameter, utility con-sumption and ejector-system size. Commercial performance of adeepcut vacuum unit operating at a HVGO true boiling point(TBP) cutpoint of 1,150°F highlights the impact of off-designejector performance on gas-oil yield. Understanding the vacuumejector-system fundamentals is critical to maintaining gas-oilyields.

Ejector-system performance at deepcut vacuum column pressuresmay be independently or concurrently affected by:

• Atmospheric column overflash, stripper performance or cutpoint

• Vacuum column top temperature and heat balance

• Light vacuum gas-oil (LVGO) pumparound entrainment tothe ejector system

• Cooling-water temperature

• Motive steam pressure

• Non-condensible loading, either air leakage or cracked light-endhydrocarbons

• Condensible hydrocarbons

• Intercondenser or aftercondenser fouling

• Ejector internal erosion or product build-up

• System vent back pressure.

Minimizing ejector-system gas loading lowers column pressure,thereby increasing gas-oil yield. By optimizing process perform-ance when processing West Texas Intermediate (WTI) crude, the

gas-oil yield can be increased by 0.75 vol%. This represents 1,150

bpd of incremental gas-oil recovery for a 150,000-bpd refinery.Assuming an average $5/bbl gas-oil differential over vacuumresidue, incremental annual revenue is $2 million. Experiencewith deepcut vacuum unit operation on WTI crude has shownthat vacuum column pressure is strongly impacted by atmospher-ic column operation and LVGO pumparound operation.

Hydrocarbon Processing, October 1994 1

Fig. 1. Gas-oil yield.

Fig. 2. Feed enthalpy vs. temperature.



GAS-OIL YIELDS

The gas-oil yield on a crude vacuum column is controlled by feedenthalpy. I f more heat can be added to the reduced crude at a givencolumn pressure, more oil is vaporized. A good furnace design isrequired to reliably meet the coil outlet temperature requirementsof a deepcut operation without excessive cracked-gas production.

Fig. 1 shows the impact on gas-oil yield, assuming a given qualityof WTI reduced crude. The curves are in terms of vacuumresidue yield as a percent of whole crude. Fig. 2 represents feedenthalpy as a function of temperature and pressure. Figs. 1 and 2are based on the same atmospheric residue composition assuminga crude unit charge of 40,000 bpd. The effect of column temper-ature and pressure on gas-oil yield is highlighted. Gas-oil yieldimprovements for small incremental pressure reductions are high-er at low column pressures than at higher pressures.

For example, a 2 mmHg pressure reduction is made for columnsoperating at 16 mmHg and 8 mmHg. Both have a constant flash-zone temperature of 760°F. Lowering pressure from 16 to 14mmHg and from 8 to 6 mmHg will increase gas-oil yield by0.46% and 0.77%, respectively. This trend is more dramatic forlarger spreads in operating pressures. The column top pressurevaried between 4 and 16 mmHg and was caused by the processand utility systems.

It is important to achieve lower pressures while meeting the practi-cal limits of furnace design and minimizing cracked-gas formation.Example: a vacuum unit is to minimize residue yield to 9% basedon whole crude. From Figs. 1 and 2 a column operating at 6mmHg and 730°F flashzone pressure and temperature will havethe same gas-oi l recovery as a column at 14 mmHg and 780° F.These two cases have a feed enthalpy differential of 171.5MMBtu/d with the higher pressure requiring a higher feedenthalpy.

EJECTOR-SYSTEM FUNDAMENTALS

Gas load.The ejector-system loading consists of:

• Non-condensibles like cracked gas from the furnace and airleakage

• Condensible hydrocarbons carried with non-condensibles

• Entrainment

• Furnace coil steam

• Tower stripping steam.

Non-condensibles and a small amount of condensible gases aregenerated in the furnace. Cracking is most severe in dry vacuum-tower operations with furnace-outlet temperatures above 750°F. Aproper furnace design will minimize cracked hydrocarbon gases.Deep-cut operations with insufficient quench to the tower bootcan also cause cracked-gas formation. The quench distributionquality to the boot should be included in the vacuum towerdesign. Ejector load is also affected by poor crude stripping in theatmospheric crude tower. Cause: damaged or an insufficient num-

ber of stripping trays, improperly designed trays or insufficientstripping steam.

Theory.The operating principle of an ejector is to convert pres-sure energy of the motive steam into velocity. This occurs byadiabatic expansion from motive steam pressure to suction-loadoperating pressure. This adiabatic expansion occurs across a con-verging and diverging nozzle (Fig. 3). This results in supersonicvelocity off the motive nozzle, typically in the range of mach 3 to4. In actuality, motive steam expands to a pressure lower than thesuction load pressure. This creates a low-pressure zone for pullingthe suction load into the ejector. High-velocity motive steamentrains and mixes with the suction gas load. The resulting mix-ture’s velocity is still supersonic.

Next, the mixture enters a venturi where the high velocity recon-verts to pressure. In the converging region, velocity is convertedto pressure as cross-sectional flow area is reduced. At the throatsection, a normal shock wave is established. Here, a dramaticboost in pressure and loss of velocity across the shock waveoccurs. Flow across the shock wave goes from supersonic ahead of the shock wave, to sonic at the shock wave and subsonic after theshock wave. In the diverging section, velocity is further reducedand converted into pressure. Fig. 3 shows ejector components anda pressure profile.

Motive pressure, temperature and quality are critical variables forproper ejector operating performance. The amount of motive steamused is a function of required ejector performance. The nozzlethroat is an orifice and its diameter is designed to pass the specifiedquantity of motive steam, required to effect sufficient compressionacross the ejector. Calculation of a required motive nozzle throatdiameter is based on the necessary amount of motive steam, itspressure and specific volume. The following equation found in theHeat Exchange Institute Standard for Steam Jet Ejectors is com-monly used to determine throat diameter:


Fig. 3. Ejector components and pressure profi le.



lb/hr motive steam = 892.4 C d D

n 2 (Psia/Vg) 0.5 where

Cd = Nozzle discharge coefficientD = Nozzle throat diameter, in.

Psia = Motive steam pressure at ejector, lbf /in2

V g

= Motive steam specific volume, ft3 /1b.

Motive steam quality is important because moisture dropletsaffect the amount of steam passing through the nozzle. High-velocity liquid droplets also prematurely erode ejector internals,

reducing performance.

Operating a vacuum unit requires an ejector system to performover a wide range of conditions. Loads vary from light to abovedesign. The ejector system must be stable over all anticipatedoperating conditions. Determinating design air leakage and light-end hydrocarbon loading is essential to stable operation of thevacuum system. Furthermore, an accurate understanding of ejec-tor-system back pressure for all operating modes is necessary forstable operation. An ejector does not create its discharge pressure,it is simply supplied with enough motive steam to entrain andcompress its suction load to a required discharge pressure. I f theejector back pressure is higher than the discharge pressure it can

achieve, then the ejector “breaks” operation and the entire ejectorsystem may be unstable.

Compression ratio. The ratio of discharge pressure to suctionpressure is the ejector compression ratio. These normally varyfrom 3 to 15. An ejector’s individual compression ratio is a func-tion of cooling-water temperature, steam use and condensationprofile of hydrocarbons handled. The first-stage ejector, tieddirectly to column discharge, will have a compression ratio setprimarily by intercondenser cooling-water temperature.

Intercondenser capital cost, steam costs and cooling-waterrequirements should be balanced against first-stage ejector designdischarge pressure. This pressure must be high enough for con-densation to occur in the intercondenser. With 85°F coolingwater, an initial steam condensing temperature of 105°F is rea-sonable. This corresponds to a first-stage intercondenser operatingpressure of approximately 60 mmHg. But, other condenser oper-ating pressures are possible. I f a lower operating pressure isconsidered, this lowers the available log-mean temperature differ-ence (LMTD) and, thus, increases intercondenser cost. But, lessmotive steam is required. I f a higher operating pressure is used,more motive steam is needed to permit compression to that high-er pressure. Capital and operating costs are balanced to optimizeoverall system cost.

A deepcut column with an operating pressure of 4 mmHg willnormally have a three-stage ejector system. Some columns havedesign top pressures below 4 mmHg and as low as 1.5 mmHg.These columns may have four-stage ejector systems. A four-stagesystem will have two ejectors in series compressing column over-head load to the first intercondenser operating pressure.

CONDENSERS

Intercondensers are positioned between ejectors. The aftercon-denser is located after the last ejector. There is an interdependencybetween the ejectors and condensers. Both must perform satisfac-torily for proper system operation. Condenser performance isaffected by:

• Cooling-water temperature, flowrate and temperature rise

• Non-condensible loading

• Condensible loading

• Fouling

• Height of barometric leg.

The first intercondenser is the largest and primary condenser inthe vacuum system. But, the second intercondenser and aftercon-denser are also key to proper overall system operation. In the past,direct-contact barometric condensers were commonly used.However, shell-and-tube condensers are primarily used now. Theycondense motive steam and condensible hydrocarbons, and coolnon-condensible gases normally on the shell side. Cooling wateris typically on the tubeside.

Configurations.The ejector system may be configured a numberof different ways to handle various crudes and differing refineryoperations. It is possible to use a single vacuum train consisting ofone set of ejectors and condensers. This allows minimal initialinvestment but l imits flexibil ity in controlling util it ies or manag-ing different crudes and varying unit operations. Often, parallelejectors are installed for each stage. Each parallel ejector will han-dle a percentage of total loading, i.e., twin-element ejectors eachdesigned for 50% of design load, triple-element ejectors eachdesigned for 40% of design load for 120% capacity, or twin-ele-ment 1 ⁄ 3 – 2 ⁄ 3 ejector trains.


Fig. 4. T hree-stage vacuum system.



Parallel ejector trains allow one train to be shut down for mainte-nance while the column operates at a reduced load. At light loads,a train may be shut down to reduce operating costs. Fig. 4 showsa typical deepcut vacuum system with triple-element ejectors and

first intercondensers. The second intercondenser and aftercon-denser are both single elements.

VACUUM-SYSTEM TROUBLESHOOTING

Commissioning. Before startup, the ejector system should be iso-lated from the column and load tested to see if air leakage occurs.Each ejector will have a “no-load” suction pressure, supplied bythe ejector manufacturer. No-load suction pressures attained inthe field should be compared to manufacturer data. If the designno-load suction pressure cannot be met, the cause should be iden-tified prior to startup.

Operation. Column overhead pressure rising above the designmaximum pressure may be the result of increased unit through-put, furnace problems or atmospheric column internal damage.These process conditions result in increased column overhead gasrate and are not necessarily a problem for the ejector system. Afirst-stage ejector will have an operating curve, usually providedby the manufacturer, that indicates column top operating pressuremaintained by the ejector as a function of mass loading. As over-head mass flowrate increases above design, so will columnoverhead pressure. The converse is also true to a point.

The ejector system will track this performance curve provideddesign air leakage or non-condensible hydrocarbon loading is notexceeded. If this happens, the first ejector will follow its perform-ance curve. However, the secondary ejectors are affected. Due toan increase in non-condensible loading, a subsequent increase insaturated condensible loading from the first intercondenserresults. As non-condensible loading increases, the amount of steam and hydrocarbon condensed decreases. The increased gasloading exiting the first condenser cannot be handled by the sec-ond-stage ejector at the intercondenser design operating pressure.Furthermore, the first-stage ejector does not have enough energy,nor are its internals designed to compress the load to a high

enough pressure to allow the second-stage ejector to handle theincreased intercondenser gas discharge. As a consequence the first-stage ejector breaks operation and tower pressure rapidly increasesand may become unstable. A similar situation may also occurbetween the second- and third-stage ejectors.

Unstable column operation can also result from poor steam con-ditions. If the steam pressure at any ejector falls below design,then less steam will pass through the motive nozzle. This results

in insufficient steam for compression across the ejector. Excessivesuperheat will have a similar effect since less steam passes throughthe nozzle. Accurate assessment of steam conditions is cri tical. I f steam pressure is below design or if excess superheat exists, themotive nozzle must be rebored to a larger diameter. After boring,it is necessary to smooth the internals and remove rough edges sothat flow coefficients are not impacted.

High steam pressures are normally acceptable as long as they arewithin 110% to 120% of design. If steam pressure is too high, thentoo much steam passes through the nozzle, choking the diffuserthroat and reducing the load handled. If this occurs, new nozzlesmust be installed or steam pressure controlled closer to design.

Wet steam causes erosion of ejector internals. This reduces ejectorcapacity and may cause erratic operation. Moisture droplets accel-erated to supersonic velocities are very erosive on the motivenozzle, inlet diffuser and exhaust elbows or condenser tubes.Steam lines must be insulated up to the ejector motive nozzle. Asteam separator and trap must be installed before each ejector.Steam traps require periodic inspection to ensure they properlydump condensate. I f the motive nozzle or diffuser show excessivewear (cross-sectional area increase in excess of 7% of design) thenthey must be replaced.

If a condenser becomes fouled, it will not properly condense andcool gases to the design outlet temperature. This increases gasloading to the following ejector, which is unable to handle thecondenser’s design operating pressure. This leads to “ breaking” of the preceding ejector. Condensers should be periodically cleanedand maintained in a usable condition.

Insufficient cooling water will similarly affect ejector performance.This problem reduces overall heat-transfer rate and increases thewater temperature rise. A higher temperature rise lowers LMTD.This effect has the largest impact on the first intercondenser. If theoverall heat-transfer rate and LMTD fall below design, condenserperformance is compromised. The net result is that proper conden-

sation and gas cooling does not occur and the ejector overloads.Therefore, the ejector system may operate in a “broken” condition.Cooling-water supply should be maintained at high enough flow tomeet the design LMTD at design heat loads.

Good condenser performance is needed the most during the sum-mer. At this time, cooling water is the warmest and refinerycooling demands are highest. Proper determination of coolingwater availability, temperature, operating pressure and pressuredrop is key to proper ejector system performance. Periodic fieldsurveys should be performed.


Fig. 5. Intercondenser tailpipe arrangement.



Improper barometric leg (condensate drain) layout has a negativeeffect on condenser performance. Condensate drains by gravity,so the barometric leg must be high enough to ensure that con-densate does not enter and flood the condenser. Flooding lowertubes makes them ineffective for heat transfer. The barometriclegs should be at least 34 ft above the condensate receiver for100% water condensate. With mixed water and hydrocarboncondensates, a barometric leg of at least 42 ft is required. If con-denser flooding occurs, check the drain legs for blockage. Also,horizontal drain leg runs are not recommended because they aresusceptible to gas pockets. Fig. 5 has recommended barometricleg layouts. I f a condensate receiver operates above atmosphericpressure then the barometric leg height must be increased.

If the system back pressure, or back pressure on any ejector,increases above its design discharge pressure, then that ejectormay operate in an unstable or broken condition. This occursbecause the ejectors’ internals are not designed to compress to ahigher discharge pressure. Also, there is insufficient steam to dothe necessary work.

PROCESS OPERATIONS

Once the ejector system is designed and the utility-system per-formance is established, process operations will dictateejector-system performance. Air leakage and non-condensibleproduction from the vacuum unit fired heater is set for a givensystem volume and furnace performance. Furnace non-condensi-ble production can be controlled by coil steam injection. Coilsteam will load the ejector system. Hence, the optimum coil

steam versus cracked gas production impacts vacuum once thesystem is built. Here, we will assume that non-condensible andcoil steam load on the ejector system are constant. Lieberman2

covered the importance of furnace design and operations on ejec-tor-system performance. We will focus on controlling the ejector-system condensible hydrocarbon load.

The condensible load is impacted by the atmospheric columnperformance and vacuum column operation. The LVGO topproduct vapor pressure has the biggest impact on ejector conden-sible load for a given ejector-system noncondensible load.Assessing the operating variables that impact LVGO vapor pres-sure is the key to minimizing it. The variables that impact LVGO

vapor pressure are atmospheric column overflash, stripper per-formance and cut-point; and vacuum column top temperatureand LVGO/HVGO material balance.

Atmospheric column. This design and operation has a significantimpact on the ejector system. For lighter material being fed to thevacuum column, LVGO vapor pressure will be higher at a giventemperature. The atmospheric column stripping section and washsection affect the vacuum column condensible load. Minimizingatmospheric column overflash is important. The stripping sectionperformance is affected by steam rate (lb steam/bbl atm. residue).Maximizing stripping-section performance is the largest, and leastcostly operating tool to maximize vacuum column ejector-systemperformance. Atmospheric column cutpoint is also important.The order of importance is:

• Stripping section

• Overflash

• Atmospheric residue cutpoint.

Vacuum column. The top temperature should always be mini-mized and the quantity of LVGO maximized. LVGOpumparound rate and return temperature must be optimizedwithin the constraints of entrainment to the overhead system and

LVGO pumparound circuit exchanger LMTD. Minimizing thepumparound return temperature will lower condensible load for agiven exchanger surface area and utility. There are some trade-offsbecause the return temperature and LVGO yield influence con-densible load.

The pumparound condenses hydrocarbons before the ejector sys-tem. The quantity of condensible hydrocarbons to the ejectorsystem is a function of the quantity of noncondensibles and thelightest-condensible material’s vapor pressure. Minimizing toptemperature minimizes first-stage ejector vapor load.


TABLE 1. SLOP OIL ASTM D86 DISTILLATION

Vol% Temperature, °F

0 133

5 19810 210

20 223

30 235

40 246

50 262

60 278

70 309

80 385

90 509

100 643

Fig. 6. Ejector system survey, high pressure.



The column packing internals and liquid distributor may beviewed as a direct-contact heat exchanger. Poor heat-transfer per-formance, due to vapor-distribution problems or poor liquiddistribution, will increase the ejector condensible load. The toppumparound system design is critical for optimum ejector per-formance. In the past, many of these distributors were sprayheaders. A modern, high quality gravity distributor will reduceentrainment and reduce ejector condensible loading.

Pumparound spray header distributors are susceptible to plug-ging, especially if no strainers are provided. Plugging results inliquid maldistribution. Even with conservative packed-beddepths, inherent packing distribution is often not sufficient torecover from maldistribution. Trayed vacuum towers are oftenrevamped with packed designs that reuse existing draw pans. Butthese draw pans are usually designed with vapor risers that are toolarge and too few in number for packed applications. Thepumparound return temperature and heat-transfer efficiency of the pumparound tower internals set the hydrocarbon vapor equi-librium. Either poor heat transfer or high pumparound returntemperatures will increase ejector load.

An improperly-designed spray header can produce sufficiententrainment to overload the vacuum system or reduce intercon-

denser heat-transfer capability by waxing the condenser tubes.The spray-header system design requires even irrigation to thepacking’s top. Nozzles to minimize mist formation are critical.The spray header design must provide a sufficient operating rangewhile not exceeding high nozzle pressure drops that produce highquantities of mist-size droplets. Our experience has shown that anozzle pressure drop of 15 psi is typically a good maximum. Thisvaries by nozzle selection.

CASE STUDY: DEEPCUT OPERATION

A new vacuum unit was designed to operate at a HVGO TBPcutpoint up to 1,150°F. One of the design flash-zone operationswas a temperature of 770°F at 12 mm Hg absolute pressure. Thevacuum ejectors were designed for an overhead pressure of 4mmHg. The vacuum overhead pressure varied after unit commis-sioning. The minimum top pressure was typically 6.5 mmHg. A2.5 mmHg reduction to achieve the design value would result in

an additional 0.8% gas-oil yield based on whole crude (Fig. 1).Optimizing an operating unit to obtain minimum overhead pres-sure is challenging. Some of the modifications implemented hadsome interesting results.

Ejector system survey. This showed that the column design toppressure could not be obtained. And a marked deteriorationoccurred at higher crude charge rates. A survey of the overheadejector system was done at a crude charge rate of 35,000 bpsd (Fig.6) and again at a charge rate of 52,000 bpsd (Fig. 7). The columnoverhead pressure was 6.5 mmH g and 14 mmH g absolute, respec-tively. The pressure surveys were conducted with an absolutepressure manometer to ensure accurate pressure readings. Non-

absolute pressure manometers are not recommended since they areaffected by changes in barometric pressure and elevation.

Process impacts. There are two approaches to troubleshooting anyprocess problem. Try something and see what happens or studythe problem. The first approach was to make a change and seewhat happens. One theory to account for the reduced perform-ance was that wax was forming on the condenser surface fromentrained LVGO. But wax was not observed during previousintercondenser inspections. An improperly-designed spray headercan produce sufficient entrainment to overload the vacuum sys-tem or reduce the heat-transfer capability of the intercondenserby waxing the condenser tubes. Then, we reduced thepumparound flowrate from 30,000 bpd to 19,000 bpd. Weobserved that:

• Top column pressure lowered by 4 mmHg to 9 mmHg

• Pumparound return temperature lowered from 125°F to 115°F

• Lower top column temperature was reduced by 6°F

• LVGO yield was the same

• LVGO draw temperature increased by 40°F

• Slop make was reduced.

Next, a study was done to determine the cause of the ejector sys-tem’s poor performance. Our initial theory was improper design.We decided to conduct a detailed survey to find out what wascausing the poor ejector performance. Evaluating unit operatingdata and looking at oil and gas samples from the overhead systemrevealed some possible problem sources. A slop oil sample fromthe ejector-system hotwell was taken and tested. Distillation datais shown in Table 1.

The distillation showed that 90% of the material was keroseneand lighter. The light material was either carried over from theatmospheric crude column or formed in the heater by cracking.


Fig. 7 Ejector system survey, low pressure.



Normally, gasoline/kerosene boiling-range material formed in theheater is minimal. The slop-oil rate was much higher than pre-dicted, based on column overhead temperature and measurednoncondensible load. Material boiling at temperatures above450°F should not have been present. The slop-oil analysis indicat-ed the atmospheric tower was not stripping and only a smallamount of LVGO was being entrained. We assumed that reduc-ing condensibles to the ejector may reduce column pressure.

An evaluation of the intercondensers was conducted, includinginstallation of block valves to isolate one of the two parallel first-stage ejectors and intercondensers for cleaning without a unitshutdown. The high proportion of non-condensible gases increas-es the difficulty of achieving low approach temperatures becauseof the relatively poor heat-transfer coefficient. The cooling-watertemperature typically ranged from 74°F to 80°F. This, in con- junction with the exchanger approach temperature limitation, setthe first-stage intercondenser pressure. The exchanger approachtemperature was about 20°F. The intercondenser performance wasadequate.

At low charge rates, the atmospheric tower’s furnace was not ther-

mally limited. Therefore, the atmospheric column cutpoint couldbe increased, lowering the light slop to the vacuum unit. At highcharge rates, stripping steam was reduced to the flooding limit onthe stripping trays. Light material to the vacuum column increased.

Reducing the light slop oil from the atmospheric column, assum-ing this caused the high condensibles load, required modificationto the atmospheric column stripping section. The vacuum col-umn light slop oil material is a result of either poor stripping inthe atmospheric crude tower or cracking in the furnace. However,furnace cracking was assumed to be negligible.

Further analysis showed that the atmospheric tower stripping sec-tion was inadequately designed. Adequate stripping steam at highcrude charge rates was not possible. The trays were hydraulicallylimited and flooded. Introducing appreciable quantities of steamresulted in black diesel oil. Tray modifications were planned dur-ing an atmospheric crude unit shutdown.

By modifying the stripping trays and improving stripping efficien-cy, slop-oil make was reduced, even at higher crude throughputs.Result: a vacuum tower overhead pressure that varied between 3mmHg to 4.5 mmHg depending on ambient temperature andhumidity. Cooling-water temperature to the first-stage ejectorintercondensers and LVGO pumparound return temperature

became the major factors in minimizing vacuum-column top pres-sure. A hydraulically-limited stripping section is not a typicalrefinery problem. But, an inefficient or damaged stripping sectionis common. When operating at low column pressure, the impactof atmospheric-column stripping-section operating inefficienciesresults in significant gas-oil yield losses due to loss of vacuum.

LITERATURE CITED

1. Kister H.K., et al., Disti llation Design , Chapter 9, McGraw-Hill Book Co., New York, 1992.

2 Lieberman, N.P. “Delayed coker-vacuum tower technology,”New Orleans, La., May 1993.

THE AUTHORS

Gary R. Martin is an independent consultant athis company, Process Consulting Services,Grapevine, Texas. His work has included fieldtroubleshooting, revamping process units andfield inspection. He previously worked as arefinery process engineer for El Paso Refiningand Glitsch, Inc. Mr. Martin holds a BS in

chemical engineering from Oklahoma State University.

James R. Linesis vice president of engineeringfor Graham Manufacturing Co., Inc., Batavia,N.Y. Since joining Graham in 1984, he has held

positions as an application engineer, productsupervisor and sales engineer focusing on vacu-um and heat transfer processes. M r. Lines holdsa BS degree in aerospace engineering from theUniversity of Buffalo.

Scott W. Golden is refinery technical servicemanager for Glitsch, Inc. in Dallas, Texas. Hespecializes in field troubleshooting, process unitrevamps and field inspection. He has publishedmore than 20 technical articles concerning refin-ery process unit troubleshooting, computermodeling and field inspection. Previously he

worked as a refinery process engineer M r. Golden holds a BS inchemical engineering from the University of Maine.






Fig. 1: Steam vacuum refr igeration system wi th barometric type

condenser

Fig 1a: A 2,000-ton vacuum refrigerati on system with barometr ic

condensers installed in a refinery



Fig. 2: Steam vacuum refr igerat ion system with sur face type condenser.

Fig. 3: Steam vacuum refr igeration system with evaporative

type condenser.



selected. In systems where the refrigerant temperature range islarge, say 45 to 65° F, a two stage flash system will result ineconomies. This can easily be understood by reference to Fig. 4. If half the load is flashed at 55° F, the steam requirement is 16.7 lbs.per hr. ton of refrigeration, and the water requirement is 5.3 gpmper ton of refrigeration for this portion of the load. The other half of the load is flashed at 45° F, requiring a steam a rate of 23.6 lbs.per hr. per ton of refrigeration, and the water requirement is 6.4gpm per ton of refrigeration. Thus the total steam consumption is

lower than if all of the refrigerant were flashed at 45° F.

In actual practice the loads are balanced in proportion to theflashed volume, but the effect of two stage flash is essentially asillustrated above.

SELECTING A STEAM VACUUMREFRIGERATION SYSTEM

• Determine desired chilled water temperature (check range)

• Determine maximum available condensing water temperature

• Determine available steam pressure.Having determined the above, you may make several selectionsusing Fig. 4, depending upon whether you wish to optimizesteam consumption or water consumption.

For surface condenser systems, condensing temperatures will beapproximately 8° above the leaving water temperatures; whereasin barometric systems, condensing temperatures will be approxi-mately 3° above leaving water temperatures.

Thus utility requirements for surface condenser systems will besomewhat above those for barometric systems.

Cost Data. Fig. 5 gives installed cost data per ton of refrigerationfor packaged barometric systems using one, two or three boosters.Fig. 6 gives installed costs for systems using surface condensers.Steam consumption of steam vacuum refrigeration systems usingevaporative condensers (Fig. 3) is a function of wet-bulb tempera-ture. Steam consumption for these systems may be determined bythe use of Fig. 8.

Cost correction factors for motive steam pressures are given inFig. 9. Fig. 7 gives cost per ton of refrigeration for systems withevaporative condensers.

Hydrocarbon Processing, June 1967 4

Fig. 4: Uti li ty requirements for steam vacuum refr igerati on

systems.

Fig. 5: Installed cost per ton, barometric systems using one,

two or three boosters.

Fig. 6: Installed cost per ton, sur face condenser systems, using

one, two or three boosters.



Multiple boostersare used for refrigeration systems requiring oper-

ation under a variable demand. Capacity control is obtained bycycling boosters in and out of service, either automatically or man-ually, with variations in load. Generally speaking these boosters aredivided in equal increments; however, should circumstancesrequire they may be subdivided in any proportion. Multiple boost-er systems are particularly advantageous if it is desired to takeadvantage of part load economy and the effect of reduced con-denser water temperature during cool periods of the year.

Multiple Systems. For sizes beyond those shown, multiple systemsare generally used; although if the size is only 10 or 15 percentlarger than shown, it is quite possible to enlarge the standarddesigns. Generally it will be more economical and give more flexi-bility to go to multiple systems. However, single units have beenbuilt as large as 2,000 tons of refrigeration.

Space requirements for the three types of systems are given in theblock diagrams, Figs. 10, 11 and 12.

Indexing Terms: Barometr ic—9, Condensers/ProcessEquipment—9, Cooling—9, Costs—7, Curves—10,Estimating—8, Evaporation—9, Heat Transfer—9, Refr igeration—9, Selection—8, Steam—6, Surfaces—9,Water—6.

Hydrocarbon Processing, June 1967 5

ABOUT THE AUTHOR

Elliot Spencer is a sales engineer with GrahamManufacturing Co., Inc., Great Neck, NewYork. M r. Spencer holds an M.S. degree inchemical engineering from the BrooklynPolytechnic Institute. He held engineeringpositions with the York Corp. and The TraneCo. prior to assuming his present position 15years ago. He is a member of ASHRAE.

Fig. 7: Install ed cost per ton for systems wi th evaporative

condensers.

Fig. 8: Steam Consumption in pounds per hour per ton

of refr igeration.

Fig. 9: Cost correcti on factors for various steam pressures.



Ejectors Give Any Suction PressureRecent tests on multistage ejector systems wi ll simplify yourtask of designi ng vacuum-producing equi pment for any pressure.

F. DUNCAN BERKELEY

GRAHAM MANUFACTURING CO., INC., BATAVIA, N.Y.

Because of overlapping performance, it’s often a lengthy prob-lem to arrive at the most economical design of an ejector. In

practically every new application of high vacuum, we find it nec-essary to investigate thoroughly the many available means of producing vacuum to reduce equipment and operating costs to apractical and profitable level.

But the giant strides of technology have brought to light an

entirely new concept in the study of vacuum-producing appara-tus. Recent tests of 5-stage and 6-stage systems indicate that

steam ejectors have carved a unique and popular place in industrywhere large volumes of gases must be evacuated—and they canproduce almost any desired suction pressure.

In addition, by using only certain parts of a multistage system,one installation can serve the whole range of test conditions.

The simple principles on which ejectors operate and the almostuniversal use of steam and compressed air in plants of all kinds

have given the ejector many advantages over other vacuumpumps. However, in spite of simple operating principles, the mosteconomical design of an ejector is often a lengthy problem.

Chemical Engineeri ng, Apri l 1957 1



Among the variables that you should consider in selecting a par-ticular design of steam ejector are:

1. Suction pressure required.

2. Steam available.

3. Water available.

4. Fluid to be evacuated.

5. Equipment cost.

6. Installation cost.

LET’S DISCUSS PRINCIPLES

In order to show how these six variables affect the design of asteam ejector, let’s discuss briefly the principles of ejectors.

All ejectors operate on a common principle. By means of a high-velocity jet of propelling steam, air or other fluid, a gas orvapor—or even finely divided solids—can be entrained andcaused to flow at high velocity along with the motive stream.

Directing the combined stream into the diffuser section of anejector converts velocity into pressure. In effect, the high-velocitycombined stream pushes against the discharge pressure of theejector and maintains a pressure difference between the suctioninlet and the discharge of the ejector.

The line sketch above illustrates approximately a typical conver-sion of pressure to velocity in the nozzle of the ejector and theconversion of velocity into pressure in the diffuser.

In all flow processes there are energy losses. The ejector is noexception.

Let’s suppose that the flow process within an ejector is 100% effi-cient. At 100% efficiency, it would be possible for an ejectorhandling no load to convert the energy of pressure of the motivegas to velocity in the nozzle and then convert this energy of veloc-ity back to pressure in the diffuser so that the discharge pressureof the ejector would equal the initial pressure of the gas.

Such ideal flow processes can be approached in a well-designedflow section, where the expansion ratio of the gas is not too great.However, the jet velocity we achieve in this instance is not veryhigh and there is relatively little velocity energy available to

entrain a secondary gas.

Under normal circumstances the expansion process in the nozzleof a well-designed ejector is almost always a fairly efficient part of the overall flow process. So we get very small energy losses in thenozzle. However, as jet velocity is increased by altering the design,the task of efficiently converting velocity back into pressurebecomes increasingly difficult. It is in this part of the flow processof an ejector that we lose some of the energy.

When we reach supersonic-flow velocities, shock waves areunavoidable in converting velocity back to pressure. These shocklosses in the diffuser become more severe as the diffuser entrancevelocity (velocity of compression) is increased. This, in turn, lim-its the discharge pressure to which the velocity energy can beconverted.

Therefore, if we fix the discharge pressure—as it is for a single-stage ejector discharging to the atmosphere—there is a practical

limit to the velocity of compression for which an ejector can bedesigned. And in the case of an ejector that is evacuating a closedvessel with no in-leakage, there is a limit to the absolute pressurethat a particular number of stages will ultimately reach, even if wepermit the ejector system to operate forever.

Suction pressure of an ejector handling a gas load is further affect-ed by the surrender of part of the energy of the jet velocity toentraining (or accelerating) the load gas. This explains why theabsolute pressure increases as the load to the ejector increases andwhy the number of ejection stages increases as the design pressuredecreases.

USE WATER TO CONDENSE

Where water is available at reasonably low temperatures, it’s com-mon practice to condense the steam from each stage of amultistage ejector in an intercondenser to reduce the load on thesuccessive stage.

Such a design reduces the steam required to handle a given loadas compared to a multistage noncondensing ejector, where eachpreceding stage discharges directly to the succeeding stage.

However, an intercondenser increases the ini tial cost of an ejectorand the problem of selection is one of operating cost vs. initialequipment cost. Because every ejector application has its owneconomics, we can’t set down a simple rule to guide the selectionof the correct design. For a particular application, though, a buyerof ejectors often knows from experience the limitations on steam,water or money that he faces.

A FAMILY OFDESIGNS

Since an ejector can

be designed for highefficiency at some par-ticular absolutepressure, each designwill yield a differentperformance curve.Fig. 1 indicates theperformance for afamily of designs of 1-stage ejectors using thesame quantity of steam in each design.




The envelope of thisfamily of curves isthe curve of all pos-sible points of maximum efficiencyfor 1-stage ejectors.If we plot manygraphs similar tothat shown in Fig. 1

for many 1- to 5-stage ejector systems,the envelopes of theindividual graphswill lead us to theoverall plot, shownas Fig. 2 on the fac-ing page.

Fig. 2 plots absolutepressure vs. air loadfor all the possiblepoints of maximum

efficiency coveringthe entire range of absolute pressures for which we usually useejectors. The data are based on ejectors designed for maximumair-handling capacity at a particular pressure and include all of the most common ejector designs based on one steam consump-tion (100-psig. steam) and condensing water at an inlettemperature of 85° F.

We can see that as many as three noncondensing stages can beused practically. In 3-, 4- or 5-stage ejectors it’s necessary to usenon-condensing stages where the interstage pressure at which acondenser would have to operate would be too low for the waterto condense the steam.

Fig. 2 permits a comparison of capacities of the various designs of ejectors that can be used for a particular suction pressure. Forinstance at 10 mm. Hg abs., four designs are available. They are:

• A 2- or 3-stage noncondensing system.

• A 2- or 3-stage condensing system.

From Fig. 2, we can see that a 2-stage noncondensing ejectorwould require about 9% more steam/lb. of air load than the 3-stage noncondensing ejector. H owever, the 3-stage ejector wouldcost considerably more than the 2-stage. Thus, there probably

would not be enough advantage at 10 mm. H g abs. to justify theadditional initial cost of the 3-stage system.

The 2- and 3-stage condensing ejectors would require only 43%and 19%, respectively, of the steam required for a 2-stage non-condensing ejector. Of course, their initial costs would be higherand they need a supply of cooling water. If long periods of opera-tion are required, however, the steam savings will undoubtedlymore than make up for the difference in initial costs.

If we know the utility and equipment costs, it’s a simple matter tocalculate how many hours of operation will be required for thesteam savings of the higher-cost designs to balance the increasedinitial equipment cost and increased cost of installation.

Installation costs can be an important consideration if steam andwater lines must be extended any appreciable distance to the ejec-tor, or if special structures must be erected to support the ejector.Ordinarily, a 1-stage ejector can be supported by the equipment

on which it is installed. However, multistage ejectors with inter-condensers require some kind of support if they are to beelevated, as they often are.

WATER TEMPERATURE EFFECTS

If condensing water colder than 85° F. were used for our compari-son in Fig. 2, all of the curves representing the performance of ejectors that require water would be shifted to the right, indicat-ing an increase in capacity for these designs.

If water warmer than 85° F. were used, the shift in these curves

would be to the left. And if the water temperature were highenough, some of the curves would move far enough to the left todisappear from the graph entirely.

The effect of water temperature is more critical on ejectorsdesigned for low absolute pressures. For example, in a 4-stageejector, the increase in capacity for 65°-F. water over 85°-F. waterfor a part icular steam consumption will be greater at 1 mm. Hgabs. than at 4 mm.

STEAM PRESSURE EFFECTS

Steam pressures higher than 100 psig. will permit designing for alarger capacity for a particular steam consumption. A greater ben-efit from high steam pressures can be realized in 1- and 2-stageejectors than in other designs.

The benefi t from high-steam pressures becomes less as theabsolute pressure for which the ejector is designed decreases.Single-stage ejectors designed for absolute pressures less than 200mm. Hg abs. cannot operate efficiently on steam pressures below25 psig. However, initial stages of multistage ejectors can often bedesigned to operate efficiently on steam pressures below 1 psig.

And it is not uncommon to use an extra stage for an ejectordesigned to operate on steam pressures as low as 15 psig.

It is very important that the steam used to motivate ejectors be atleast dry-saturated steam. Small amounts of moisture can beremoved successfully by using a good, properly sized steam sepa-rator which will remove 98 to 99% of the moisture entering theseparator. Moisture in steam is usually difficult to detect withoutthe careful use of a throttling calorimeter. Steam calorimeters arelaboratory instruments and are seldom available in the field.






Many an engineer has had difficulty proving or disproving thatthe quality of steam is affecting the operation of an ejector.

Steam separators are relatively inexpensive and should always beinstalled with an ejector wherever there is any possibility that thesteam to the ejector contains moisture.

Steam lines from the boiler to the ejector should be insulated—especially where the length of piping is over 10 ft.—because if aboiler is generating steam that is just barely dry-saturated, it willtake a relatively small heat loss to cause moisture to be present inthe steam at the ejector.

WHY USE INTERCONDENSERS?

Condensing ejectors are available in both surface or barometric(direct-contact) types.

We have not shown the economic considerations of water require-ments on Fig. 2, but we should mention that the barometricintercondenser requires slightly less water to operate than the sur-face intercondenser.

Barometric intercondensers have these principal advantages:

• They cost less than a surface intercondenser designed for thesame service.

• If used with a barometric leg, they don’t need a condensate pump.

• They seldom require cleaning and can handle corrosive ortarry substances with relatively little wear or loss in efficiency.

• The vapors come in intimate contact with the condensingwater in a scrubbing action that removes soluble vapors,gases and suspended solids from the noncondensables.

The disadvantages of barometric intercondensers are:

• Condensate mixes with the cooling water andcannot be recovered for use as hot, pure boiler feedwa-ter.

• If a pump, instead of a barometric leg, is usedto remove the water, it must handle the condensingwater in addition to the condensate. This requires alarger condensate pump than for a surface intercon-denser.

HOW TO SELECT EJECTORS

By using Fig. 2 we can make the correct selection of asteam ejector to handle noncondensable gases. In caseswhere a portion of the load to the ejector is a condens-able vapor, the data plotted on Fig. 2 are not applicableand it’s necessary to analyze the particular operatingconditions to determine the correct ejector design foroptimum economy.

In some instances we can reduce the load to the ejectorconsiderably by using a precondenser to condense a

large portion of the vapors before they reach the ejec-tor. Often the absolute pressure is too low to use a precondenserand it’s necessary to compress or boost the vapors to a pressurewhere a large part of the condensing can be done in an intercon-denser. This permits the use of small secondary ejectors tocomplete the compression of non-condensable vapors.

For a multistage ejector handling air or other noncondensablegases, there is a particular design that will require a minimum of steam and water for its operation. Using more water will not giveany appreciable steam savings.

In cases where a large portion of the load is a condensable vapor,there is a range of steam and water combinations which can bedesigned for and the relative costs of steam and water will deter-mine the best design. The cost of ejector equipment will usuallynot vary appreciably within the range of steam and water require-ments possible. So the problem in these instances is one of economics of operation where the initial cost of the ejector equip-ment is fixed.

Performance of ejectors operating on fluids other than steam can-not be analyzed by using Fig. 2, since the thermodynamicproperties of the motive fluid will vary the design of an ejector.

OPERATING CHARACTERISTICS

Each stage of a multistage ejector has the same basic operatingcharacteristics as a 1-stage ejector. Therefore, to understand theoperation of a multistage ejector, we should first discuss how 1-stage ejectors operate.

Single-point design ejectors are most sensitive to changes in dis-charge pressure. I f the discharge pressure of an ejector exceeds itsminimum stable discharge pressure, the operation will become






each condition. In changing operations from one condition to theother, we only have to shut down the system long enough tochange the nozzle or the diffuser.

Often, substantial steam savings can be realized without buyingtwo ejector systems.

Designs of this kind have found applications in the recompressionboosters for evaporators and large ejectors for high-altitude wind

tunnels.

In certain applications ejector is required to meet a specific designcurve. Then we sometimes must use considerably more steamthan for a single-point design to produce the desired characteristiccurve. At some point in the curve the ejector is, of course, rela-tively efficient and at either side of this high-efficiency point theejector is relatively inefficient.

Ejectors of this type are used frequently by jet-aircraft-engine testlaboratories where altitude conditions are simulated in a vacuumtest cell. These test cells permit us to observe and measure the per-formance of an engine under the actual conditions that it will meet

in the sky. Enormous ejectors have been built for various enginemanufacturers to handle the combustion products from a jet engineat vacuum corresponding to altitudes as high as 100,000 ft.

At these altitudes the absolute pressure dwindles to 8 mm. Hg orless. Ejectors designed for these applications must cover a wide rangeof operating conditions with a minimum steam consumption.

Fig. 3 shows typical performance curves of some large ejectorsnow being used by aircraft companies to test engines at altitudesfrom sea level to 40,000 ft.

USE ONLY SOME STAGES

It’s possible to meet a large variety of operating conditions eco-nomically with multistage ejectors by operating only some of thestages at a time.

All ejectors have at least as many different performance curves asthey have stages. For a particular stage to operate, all the succeedingstages must, of course, be operating also. Fig. 4 indicates a set of performance curves for a typical 5-stage ejector. By furnishing suit-able automatic controls, practically all points within the envelopeformed by these curves can be reached by the ejector. Thus, the

ejector can cover an entire area of possible operating conditions.

On large ejectors, the cost of automatic controls may be paid formany times in steam savings.

Six stages of compression have lengthened the range of operationof steam ejectors down to absolute pressures as low as 5 micronsof Hg (0.005 mm. H g). Commercial designs are available andshould often be used in place of other kinds of vacuum pumps.

Chief advantages of ejectors over other kinds of vacuum pumps are:

• Rugged and simple construction.

• They can handle enormous volumes of gases in relativelysmall sizes of equipment.

• Require less maintenance.

• Simple operation.

Other considerations, of course, may outweigh these advantages.Or perhaps the unavailability of a suitable motive fluid or waterwill rule out the use of an ejector for a particular application.

You’ll need an overall picture of your requirements and utilities toselect the best vacuum pump for your needs.

ACKNOWLEDGMENT

We gratefully acknowledge the comments and suggestions of H . M.

Graham and the engineering department of Graham Manufacturing Co.




The plot in Figure 4 illustrates pilot plant ejectors. In this plot,note that two- and three-stage condensing units have been elimi-nated, shutoff pressures of various other units have been altered,and a four-stage noncondensing curve has been added.

In order to illustrate the differences between the application of jetsto pilot plants and to full-size facilities, suppose one is choosing anejector for operating at 20 mm. Hg abs. For a pilot plant, onewould select a two-stage noncondensing ejector. However, as can

be seen from the curve, if it were being selected for a productionplant, one would probably choose a two-stage condensing system.

Further, if one were selecting an ejector to run on a pilot plant at5 mm., a two- or possibly three-stage noncondensing jet wouldbe used. In reference to the production plant chart, note thateconomy requires the use of a three- or perhaps four-stage con-densing jet, the four-stage unit being more economical to run.

Should one desire a 1-mm. system, one would use a three- orfour-stage noncondensing ejector on the pilot plant. For the pro-duction plant, one would need a four-stage condensing unit.

Many processes today are being investigated at 500 micron orlower in absolute pressure. For the small loads that would beencountered in the pilot plant, a four-stage non-condensing unitmay do, yet the curve shows that for the production plant a four-or five-stage condensing unit would be required.

When an ejector is required for pilot plant operation below 500micron, it normally becomes necessary for the ejector manufac-turer to supply intercondensers of some type. For the pilot plant,the direct contact, or barometric condenser is probably the mostsatisfactory, since it is the most trouble-free condenser. However,if conditions require that the condensate be recovered, the surface

type is necessary. When surface-type units are selected, theyshould be of such a design that the process side may be readilycleaned and inspected.

These considerations all revolve around economy. For small ejec-tors handling pilot plant loads, it is possible to supply a piece of equipment with a low first cost and a reasonable steam consump-tion, yet if one were selecting a unit sized for, say, ten times thecapacity, one would approach the problem in a different manner,

since the cost of steam used always amounts through the years tofar more than the original cost of the equipment. In the pilotplant, the ejector is not used very often over long periods of time.Therefore, steam consumption is not an essential considerationwhen selecting an ejector, and the non-condensing type is nor-mally recommended. Though it may lack economy, thenoncondensing type is relatively inexpensive and extremely sim-ple. Since it requires no condensing water, it offers anotheradvantage in that there is no problem of condensate removal. Itcan be mounted at ground level, as opposed to the required baro-metric leg on the condensing type with the barometric condenserand, of course, when the surface-type intercondenser is used, acondensate removal pump, or other means, is required to drain

the intercondensers.

MATERIALS OF CONSTRUCTION

In the selection of ejectors for pilot plants, it is recommendedthat alloys or corrosion-resistant materials always be selected. Thisrecommendation is made for the following reasons:


Figure 2. Three-stage condensing air ejector (right) and three-stage noncondensing air ejector (left ).



1. Many times the pilot plant will be used for only a short peri-od of time, and if the ejector is purchased as manufacturedfrom one of the more corrosion-resistant alloys, it can beused again at another location.

2. This type of ejector is so small that the alloy from which it ismanufactured has relatively little to do with the overall price.

3. Many times it is impossible to determine under what condi-tions the pilot plant will operate, and it is unknown just

exactly the service to which the ejector will be subjected. Analloy gives the best protection for this situation.

In order to give some idea as to the installed costs of ejectors,Table 1 shows estimated installed costs for production plant ejec-tors. In Table 2 are estimated installed costs for pilot plantejectors. These estimates are based on the ejectors shown inFigures 3 and 4 and indicate ejectors which use approximately1,000 lb./hr. steam and have capacities as shown.

The curves in Figures 3 and 4 are for a noncondensable load. Forunits which have a condensable and noncondensable load com-bined, these curves will not apply and may be much different.

Figure 5 shows a portable pilot plant ejector which many processplants find expedient and extremely economical. A four- or five-stage ejector has been selected for a nominal capacity and has atypical operating curve as indicated in Figure 4. It has been select-ed for the largest probable load that the pilot plant will have andis so arranged that it can be operated as a five-stage ejector withits characteristic curve, a four-stage ejector with its characteristiccurve, etc., down to a single-stage unit. The unit is self-containedand has found wide use in plants where many different pilotoperations are run in a short time period.

OPERATION AND MAINTENANCE

There are a few rather simple rules to follow in the operation andmaintenance of any ejector equipment, and if the operator willadhere to these rules, little or no difficulty may be expected.

1. It is essential that the joint between the steam nozzle and thesteam chest be tight so that there are no steam leaks at thispoint. A steam leak at the back of the nozzle acts like anadditional load on the ejector and will tend to decrease thevacuum that this apparatus can produce.

2. Be sure that steam is supplied at the design pressure and tem-

perature. Lower steam pressures cannot be tolerated underany circumstances on most ejectors, and higher steam pres-sures cause them to use more steam with no increase incapacity. Best results are obtained when the operating pres-sure is held as close as possible to the design pressure.


Figure 3. Plot of producti on plant ejectors.

Table 1.Installed costs of Figure 3 production plant ejectors usingbarom etric-type condensers (approxim ately 1,000 lb./hr.steam ),w ith

estim ates for other capacities.

Ejector Figure 3 2X 4X 1OX

Single-stage,noncondensing $ 1,200 $ 1,400 $ 1,850 $ 2,800

Two-stage,noncondensing 1,800 2,200 3,150 4,700

Three-stage,noncondensing 2,400 2,900 4,150 6,500

Two-stage,one-condenser 4,200 5,800 8,200 12,600

Three-stage,tw o-condenser 6,600 8,700 12,400 18,800

Three-stage,one-condenser 5,800 7,900 10,800 16,600

Four-stage,tw o-condenser 10,600 13,400 19,500 29,800

Five-stage,tw o-condenser 16,300 23,800 34,000 53,600

Six-stage,tw o-condenser 19,200 27,300 39,200 60,600

Note :The above figures are based on the load being totally noncondensable and wilnot apply when a mixture of condensable and noncondensable vapors is present.Multistage ejectors are based on nominal suction pressures,and figures will vary slightly for higher or lower pressures.



3. Keep the nozzles clean. It will be found that when a new sys-tem is started pipeline chips and other foreign matter arecarried in the steam lines to the ejector strainer and that cer-tain particles may pass through this strainer and plug up thesteam nozzle. This will show up in loss of vacuum. It is advis-able to blow out the strainer frequently upon first starting upand, if necessary, to check the nozzles by removing the plug atthe back of the steam chest and passing the proper reamerthrough each nozzle to make sure it is not plugged.

4. The steam supply piping to the ejector should be of suffi -cient size to pass the steam required by the ejector with noappreciable pressure drop. The steam supply piping shouldalso be short enough to assure design operating pressure. Theejector will operate most efficiently on dry steam; thus, thesteam supply piping should be insulated to prevent excessivecondensation before the steam reaches the ejector. Shortnessof steam pipe will also reduce condensation. If there is anydoubt as to whether the steam is dry, a moisture separator

should be installed in the line.5. If the unit has an intercondenser or aftercondenser of the

surface type, the tubes should be kept clean on the waterside. When these tubes foul up, they will fail to transfer suffi-cient heat to condense the steam, in which case steam willdischarge to the next-stage ejector or to the air vent of theaftercondenser. In the case of an intercondenser, this, of course, means loss of vacuum.

6. The ejector should be placed as close as possible to the vesselwhich is to be evacuated to minimize pressure drop.

It should be emphasized that the steam jet ejector is one of the

most foolproof, trouble-free pieces of apparatus that operate inany vacuum cycle. This does not mean that the apparatus can beabused beyond all limitations, nor does it mean that it can beignored indefinitely, insofar as inspection, maintenance, andrepair are concerned. It simply means that it is one of the mostdependable sources of vacuum that can be purchased.


Table 2. Installed costs of Figure 4 pilot plant ejectors (approximate-

ly 1,000 lb./hr.steam),with estimates for other capacities.

Ejector Figure 4 0.25X 0.5X 2X

Single-stage,noncondensing $ 1,200 $ 600 $ 850 $ 1,400

Two-stage,noncondensing 1,800 860 1,250 2,200

Three-stage,noncondensing 2,400 1,200 1,800 2,900

Four-stage,noncondensing 3,200 - 2,950 4,100

Four-stage,two-condenser 10,600 5,500 7,100 13,400

Five-stage,two-condenser 16,300 8,600 11,100 23,800

Six-stage,two-condenser 19,200 - 14,600 27,300

Note :The above figures are based on the load being totally noncondensable and will not apply when a mixture of condensable and noncondensable vapors is present. Multistage ejectors are based on nominal suction pressures, and figures will vary slightly for higher or lower pressures.

Figure 4. Plot of pilot plant ejectors.



Documents

eyectors.pdf