Biermann Ch. Handbook of Pulping and Papermaking

Подождите немного. Документ загружается.

634 29. NONWOOD FIBER USE IN PULP AND PAPER

phloem consists of fibers, sieve tube elements,

companion cells, and parenchyma. In monocots

these bundles are scattered in the stem throughout

the ground tissue, whereas in dicots the bundles

are all about the same distance from the center and

form a ring of bundles around the pith. In dicots,

the bundles unite as the stem grows outward; in

many species the bundles are no longer distinct

(Fig. 29-2). The scattering of the xylem remains

in those monocots that undergo secondary growth.

The outer layer of cells is the epidermis.

The layer just inside the epidermis is the cortex.

The layer inside the cortex but outside the vascular

tissue is the pericycle. Epidermal cells have

sinuous or toothed margins and may have projec-

tions (in cells called trichomes) that are helpful in

identification of samples.

Useful fibers

Useful fiber (that provides the strength for

paper made from the pulp) is derived from the

vascular tissue of monocots of barley, Hordeum

spp.;

rice, Oryza spp.; esparto, Stipa tenacissima;

wheat, Triticum spp.; bamboo, Phyllostachys(Fig.

29-3);

sugar cane, Saccharum officinarum; and

others.

Bast, fibers from the phloem of dicots, is

derived from hemp, Cannabis sativa;

kenaf.

Hibiscus cannabinus; flax, Linum usitatissimum

(Fig. 29-3); and others. Fiber may obtained from

the pericycle or cortex of some dicots. Cotton is

the seed hair of the cotton plant (Fig. 29-3). Fiber

from the vascular tissue of leafs is obtained from

sisal (Agave sisalina) and manila hemp {Musa

textilis).

Nonwood

fiber

identification

A variety of nonwood fibers can be identified

by a series of specialized stains. For example,

several are available for hemp and flax. Acidified

potassium dichromate swells flax somewhat faster

than hemp. Cyanine stains these materials differ-

ently. Other separations exist for cotton, linen,

and wood; cotton, flax, jute, and hemp; animal

and plant fibers; etc. (Standard T 401 and others).

Straw

morphology

considerations

Fig. 29-4 shows a cross section of the straw

stem. The usable fibers are the dark fibers near

the edge of the stem consisting of phloem and

other tissue.

According to Huamin (1988), parenchyma

cells occupy 38% of wheat straw based on the

cross—sectional area of the stem, so the mass

percentage may be lower. These cells are small

and thin—walled, contribute to decreased pulp

freeness, and do not add much to the strength of

paper. They cook more slowly than the fiber cells

and use more cooking chemicals. The lignin,

cellulose, and xylan composition of the various

cell types is very similar.

Petersen (1991) found the oj—cellulose

content of the internode of straw of four grains to

be 37—42% while the leaves had only 28—30%

a—cellulose. The fibers are about 15—20%

longer in the nodes compared to the leaves. The

author suggests that whole straw should not be

pulped anymore than one would pulp trees with

leaves and all. The internode is 35% of orchard

grass (including the inflorescence) and 43 to 48%

of many other grasses.

Fig. 29-2. Stem cross sections of Cannabis (left,

100

x) and Linum (right,

x).

INTRODUCTION 635

Fig. 29-3. Fibers of bamboo (left, with parenchyma, 60x), linen (center, 150X), and cotton seed

hairs (right, 150x).

Depithing

Bagasse, and some other nonwood fibers,

must be depithed before use. The pith contains

small cells (parenchyma) that do not add to the

pulp strength and reduce the freeness considerably.

29.2 PULPING METHODS FOR

NONWOOD FIBERS

Introduction

Annual plants have been used since the early

1800s, except for cotton, which has been used

much longer, of course. Straw has traditionally

been pulped by boiling solutions of lime for board

grades of paper; this led to a bright yellow pulp.

Sodium hydroxide was used to make bleachable

grades of pulp as well as board grades. The soda

anthraquinone (AQ) method is replacing the soda

method and the results of the soda/AQ for straw

are said to be similar to those for kraft pulping of

straw. Other methods of pulping include neutral

sulfite and chlorine systems.

Alkali, chlorine

The Pomilio process (Stephenson, 1951) was

the most popular method at one time, being used

in 20 mills. Developed around 1925, this process

involved treating pulp with 8% NaOH (on dry

pulp) in 60 ft cylindrical reaction towers where

temperatures up to 130°C could be reached at the

bottom. The digestion time was about 90 min. A

chlorination step follows. The overall yield is

36--40%.

The high use of chlorine and cellulose

hydrolysis make this method obsolete.

636 29. NONWOOD FIBER USE IN PULP AND PAPER

Fig. 29-4. Stem cross section of wheat (with

part of a leaf) (50x).

Sulfite

Stephenson (1951) discussed a neutral sulfite

(NSSC) method that enjoyed a lot of use in Eu-

rope.

Sodium sulfite (10%) and 5% NaOH on pulp

are combined with straw for a 6 hr cook at 160°C.

The pulp yield is about 55%. Hypochlorite (5%

on pulp) bleaching reduces the yield to 42%.

Aronovsky (1948) suggested an NSSC pro-

cess for pulping wheat straw. This process uses

8% sodium sulfite and 2—3% sodium carbonate

(on dry straw), with a liquor to straw ratio of

7:1,

and a cooking time of two hours at 170°C. The

pulp yield is 52—55% and can be brightened to

70%

with 5—7% total chlorine. Bleaching above

this brightness requires a three—stage process.

The freeness was said to be 400—500 mL.

Aronovsky (1949) found that the neutral sulfite

process was unsuitable at atmospheric pressure, in

agreement with recent unpublished results, but

atmospheric pulping for 1 hour with 12% kraft

chemicals in a hydrapuler at 90—98° was very

effective. In other work (1947), he also suggested

lime and 2% sodium sulfite to give an excep-

tionally free pulp, although it would seem that

calcium sulfite might precipitate.

Acid sulfite methods of pulping straw give

poor results due to the relatively labile carbohy-

drates of straw. Relatively weak and brittle pulps

result, and the process is unsuitable for materials

with high silica contents.

Soda

England used the soda process to make much

of its straw pulp at one time. In 1945, the United

Kingdom pulped nearly 350,000 tons of straw for

paper (FAO, 1952). Esparto uses less alkali and

gave a higher yield than the cereal straws. Typi-

cally, esparto straw was pulped with 12% NaOH

(on straw) for 5 hr at 160°C. Hypochlorite

bleaching gave a yield of 36—40%.

Anon. (1948) discussed the French process

developed by Huguenot for pulping straw. The

continuous process involves pulping chopped straw

with NaOH at 90-100°C for 30—50 minutes.

The pulp then goes to a digester pulper at 80°C,

where it is agitated for 4—7 hours. This process

is also said to be suitable for waste paper. More

information on soda pulping on nonwood plants is

found on pages 47—50.

Alkali—oxygen,

NACO

process

Ceragioli (1975) describes alkali—oxygen

pulping of straw. The author claims that sodium

carbonate does not reduce the strength of pulp as

does sodium hydroxide in the presence of oxygen.

With carbonate, magnesium addition to protect

carbohydrates from degradation appears to be

unnecessary. A sodium carbonate—based pulping

system has a simplified chemical recovery system.

The NACO process (U.S. patent 4,612,088

issued in 1986) uses oxygen and alkali (sodium

carbonate and some sodium hydroxide) to pulp

nonwood fibers (and upgrade secondary fiber) on

a relatively small scale (Anon., 1984). A continu-

ous,

pressurized reactor (Turbo—Pulper'^'^) was

developed as part of the process. The process has

been used commercially since 1986 in Italy at a

100 ton per day mill (IPZP Foggia) for nonwood

pulp and 50 tons per day for upgrading OCC.

Unbleached straw pulp has a brightness of

50—52 (ISO) and kappa number of 15—16 with a

yield of

48%.

The brightness can be increased to

72%

with one hypochlorite (H) stage or 82% with

PULPING METHODS FOR NONWOOD FIBERS 637

two H stages (6—7% as active chlorine). Ozone

bleaching gives a brightness of

75—78 %.

Pulping

is carried out at 7—8% consistency, 90 psig, and

130—135 °C for a minimum of one hour.

29.3 CONSIDERATIONS FOR NONWOOD

FIBER USE

Ash and silica both interfere with the pulping

(and chemical recovery of pulping chemicals) of

many nonwood fibers. The cellulose, hemicellu-

lose,

lignin, and extractive contents are fundamen-

tal to pulping. Silica is present at elevated levels;

while wood has less than 0.1%, many annual

crops such as straws have 0.5%—5% or more.

The chemistry of

silica

An introduction to the chemistry of silica will

serve as a basis for the behavior of silica during

pulping operations. Silicon has an atomic weight

of 28.09 and has a valence of 4 in the oxide

forms.

Silicon dioxide

(SiOj,

also called silica or

silicic anhydride) is 46.75% silicon and occurs in

nature as a variety of minerals such as the quartz

minerals and cristobalite. The Si—O bond is

partially ionic (about 50%). SiOj occurs in crys-

talline and amorphous forms. The crystallized

form is said to be inert to alkali.

SiOj is insoluble in acids and water. It is

attacked by HF and ultimately converted to the

volatile gas SiF4. (This is the basis of some straw

pretreatments with HF for silica removal.) The

amorphous form is solubilized in alkali as the salts

of silicic acid. Silicic acid has the formula of

HjSiOj and is the basis of silica gel desiccant and

opal. The soluble ion under alkaline conditions is

SiOa^- or [Si02(OH)2]2-.

Dean (1985) lists the two acid ionization

constants of silicic acid and the solubility product

of CaSiOj at 25

°C.

The pK^ for silicic acid is

9.77 and the pA^ is 11.80. (Methods that precipi-

tate silica from black liquor using COj must

reduce the pH to about 9.0 to obtain

near—complete precipitation of silica.) The p^^p

for CaSiOj is 7.60; ^sp = 2.5 X 10-^ corre-

sponding to a solubility of 0.0184 g/L.

The chemical composition of straw

Aronovsky et al. (1943) analyzed a variety of

straw materials as well (Table 29-1). The mois-

ture contents were 6.6—8.4%. Barley is known to

have a relatively high content of

pectins

and gums,

making a high water extractives level, although

others have reported much lower results for water-

soluble materials. Among the straws, the low ash

content, high cellulose content, and long fibers of

rye grain make it particularly useful for pulp.

Mineral

composition,

especially silica

Delga (1947) investigated the mineral (and

elemental) composition of cereal stalks. They are

high in soluble ash (1.54—4.7%). Potassium was

0.77—3.44% and silicon (not silica) was 0.35 to

0.73%.

The highest level of sodium occurred in

oat straw (0.85%), while the highest level of

calcium was 0.26% in rye grain straw. Sulfur and

phosphorus contents were low. Oat straws grown

in two different soils did not vary in composition.

Fahmy and Fadl (1958) found Egyptian wheat

straw to contain 8.34% ash and 3.44% silica. The

leaf (sheath) is more concentrated with 14.9% ash

and 6.90% silica while the stem is 5.90% ash and

2.24% silica. Wet or dry sorting of the

prehydrolyzed raw material was said to remove

the silica—rich epidermis with a 4% loss of

material. Some work indicates that ethanol—ben-

zene extraction allows silica to be removed with a

second extraction step.

Silica is about 2.5—3.5% in bamboo; it is

more concentrated in the nodes, where it may

occur as nearly pure silica, than in the internode

region. Bagasse is about 1.5% silica. The silica

content of straw leaves is higher than that of the

stems.

Table 29-1. Chemical composition (%) of cereal

straws after Aronovsky (1943).

Extractives

EtOH—ben2

Cold water

Hot water

NaOH

Lignin

Pentosans

Qf-cellulose

Ash

Nitrogen

Rice

:. 4.6

10.6

13.3

49.1

11.9

24.5

36.2

16.1

0.6

Barley

4.7

16.0

16.1

47.0

14.5

24.7

33.8

6.4

1.1

Wheat

3.7

5.8

7.4

41.0

16.7

28.2

39.9

6.6

0.4

Rye

3.2

8.4

9.4

37.4

19.0

30.5

37.4

4.3

0.7

Oat

4.4

13.2

15.3

41.8

17.5

27.1

39.4

7.2

0.5

638 29. NONWOOD FffiER USE IN PULP AND PAPER

These facts point out that if straw can be

preprocessed to remove leaves and nodes, then

many advantages may be realized.

Pulping and silica content

Fahmy and Fadl (1959) determined that the

duration of alkaline pulping of wheat and rice

straws was the most important parameter of the

ash and silica content of the final pulp. Other

variables have very little influence. For example,

rice straw cooked at 150°C for 0.5 hour was

lower in ash than that cooked for 4 hours, with the

effect more pronounced for leaves than for stalks.

This may partly be due to a lower pulp yield with

increased cooking time. Bleached pulps from rice

stalks cooked at 120°C for 0.5 hour had 0.064%

silica and 0.15% ash.

Sodium sulfite pulping of bamboo with kraft

green liquor removes less than 10% of the silica

from straw compared to the 60—70% removed

during kraft pulping. The sulfite process leads to

high—strength and high—yield pulps.

Alkali chemical recovery and silica

Black liquor from straw pulping has about

5500 Btu/lb (on solids) compare to 6600 Btu/lb for

black liquor from wood. A lower residual alkali

(about 3 to 4 g/L) causes lignin and silica to

precipitate during liquor concentration, making

scaling and high viscosities big problems. Liquor

viscosity is much higher than with wood pulp. A

long time ago, Rinman developed a technique of

adding some Ca(0H)2 to the pulping liquor so that

calcium silicate would precipitate onto fibers

during the cook.

Grubshein (1961) pointed out that during

causticization some of the sodium silicate is con-

verted to insoluble calcium silicate. This decreas-

es the causticizing efficiency and increases the

lime mud volume. He concluded that the answer

is removal of silica from the black liquor by one

of two methods: 1) treat the black liquor with

lime or 2) treat the black liquor with flue gases to

lower the pH and precipitate silicic acid. Method

1 is covered in the patent by Gruen (1953), who

suggested

the

precipitation with CaO occur near or

above the boiling point of the black liquor for a

short period of time (5—10 min) to decrease the

amount of organic material precipitated. The

process was patented in Germany by Schwalbe

(1929).

(One mill in South Africa uses ferric

oxide and alumina to precipitate silica.) The

CO2 method is more common since costly lime is

not used. Also, scaling will be lower in subse-

quent evaporation. The use of CO2 precipitation

with kraft liquors would probably increase the

TRS emissions in the recovery boiler since this

process is akin to direct contact evaporation.

The experimental findings of Lengyel (1960)

indicate that 6 to 8 g/L silica in black liquor can

be evaporated without corrosion provided that

addition of excess NaOH is used to dissolve

incrustations that form. Silica at this level does

not overly interfere with causticization but does

increase the quantity of lime mud by about 100%

and begins to retard the sedimentation rate.

Corrosion and scaling are aggravated by allowing

the black liquor to stand motionless for prolonged

periods of time. Black liquor with more than 8

g/L silica should be treated with CO2 (favored) or

CaO (less favored). Lime mud can be purged if

calcium silicate builds up in the system.

Sawheny (1988) reports that silica precipitates

from soda black liquor from pH 10.2 to 9.1. At

the lower pH large amounts of lignin also precipi-

tate.

The solubilities are also temperature depen-

dent. In pilot plant tests involving several species

of nonwood plants, careful carbonation (with a

bubble reactor) of black liquor containing 6 g/L

silica followed by filtering in a filter press resulted

in 90% of the silica being removed as a solid

containing 70% silica. The author claims rapid

precipitation/sedimentafion of large silica particles.

Ibrahim (1988) indicates that precipitation of

silica from black liquor is most effective if the

black liquor is preconcentrated to at least 8%

solids. Either CaO or CO2 would precipitate over

95%

of the silica under these conditions. (The

concentration is often about 4% solids off the

brown stock washers.) Long settling times (6

hours) were needed. The long time period often

means that the pH drops and lignin may continue

to precipitate. The precipitate should be washed

to recover usefiil alkali. Centrifugal separation of

the silica precipitate was the most efficient precipi-

tation method. The best precipitation of silica was

achieved at pH 9—10 (at 50°C) with a flue gas

(with CO2 concentration of 6—8%) application of

50—150 m^ per m^ of black liquor.

CONSIDERATIONS FOR NONWOOD FIBER USE 639

mM.

Fig. 29-5. Wool (left) and silk (right) fibers

(600

x).

29.4 OTHER FIBER TYPES

Animal fibers (Fig. 29-5)

Wool and human hair fibers have scales on

the surface. Silk is formed by the secretion of

proteins from two main glands of the silkworm

caterpillar (Bombyx mori). Silk often can be seen

as two separate component strands and it is usually

of uneven diameter.

Polymer fibers

Synthetic fibers (Orion and Dacron

are

shown

in Fig. 29-6) can be quite long.

Fig. 29-6. Orion and Dacron fibers (150x).

29.5 ANNOTATED BIBLIOGRAPHY

Norwood fibers

Anon., Continuous manufacture of pulps

from annual plants, Papeterie 70(9):271,

273-276 (Sept., 1948).

Anon., Italian mill tires new pulp process,

PP/(Jan.):48, 49, 52(1984).

Anon., Tappi Standard T 259

om-83,

Species

identification of nonwood plant fibers, lip.,

5 references. This reference has numerous

micrographs. It has information on the fiber

types obtained from species of commercial

importance.

Anon., Tappi Standard T 401 om-88. Fiber

analysis of paper and paperboard, 12 p., 32

references. This method describes prepara-

tion of slides with the use of stains for the

identification of fibers from wood and

nonwood plants.

Aronovsky, S.I., G.H. Nelson, and E.G.

Lathrop, Paper Trade J, 117(25):38-

48(1943).

6. Aronovsky, S.I., A.J. Ernst, and H.M.

Sutcliffe, Pulping with sodium sulfite to

produce strawboard, Tech, Assoc, Papers

30:321-323 (June, 1947).

640 29. NONWOOD FIBER USE IN PULP AND PAPER

Aronovsky, S.I., G.H. Nelson, A.J. Ernst,

H.M. Sutcliffe and E.G. Lathrop, Paper

Trade J.

127(11):

154-162(1948) and Tech,

Assoc. Papers 31:291-299(1948).

8. Aronovsky, S.I. and E.G. Lathrop, A new

mechano-chemical process for pulping agri-

cultural residues,

Tappi

32(4):145-149(April,

1949).

9. Atchison, J.E., World capacities for

nonwood plant fiber pulping increasing faster

than wood pulping capacities,

Tappi

Proceed-

ings, 1988

Pulping

Conference,

pp 25-45.

10.

Gasey, J.P., Pulp and Paper:

Chemistry

and

Chemical Technology, 2nd, Ed., Vol. 1,

Interscience, New York, 1960. Pages 398-

425 discuss the pulping of various nonwood

fibers.

11.

Geragioli, G., Wheat straw pulping by alkali-

oxygen processes—cooking variables and

pulp processes, pp. 227-252 (1975).

12.

Dean, J. A.,

Lange 's Handbook

Chemistry,

13th ed., McGraw-Hill, New York, 1985.

13.

Delga, J., Ghemical composition of the stalks

of cereals, Mem. services

chim.

etat (Paris)

33:7-73 (1947).

14.

Fahmy, Y. and M. Fadl, Digestion of wheat

straw for the economical production of puri-

fied pulps of low silica content, Textil-

Rundschau 13(12):709-719(1958).

15.

Fahmy, Y. and M. Fadl, Solubility of ash

and silicic acid during alkaline pulping of

cereal straws. Das Papier 13(13/14):311-

314(1959).

16.

FAG,

Tropical Woods and Agricultural

Resi-

dues as Sources of

Pulp,

Rome, Italy 1952,

190 p. Sulfite pulping of Parana-pine (23

p.),

chemical pulps from Australian eucalyp-

tus (6 p.), and agricultural residues (47 p.)

are some highlights of this work.

17.

Grubshein, B.D., Recovery of lime in the

alkaline pulping of straw and reed, Bumazh.

Prom.

36(8):

12-13(1961).

Russian.

18.

Gruen, B. H. E., U.S. patent 2,628,155

(Feb.

10, 1953).

19.

Huamin, Z., Separation of fibrous cells and

parenchymatous cells from wheat straw and

the characteristics in soda-AQ pulping, 1988

International

Non-wood Fiber Pulping and

Papermaking

Conference,

Ghina,

469-478.

20.

Misra, D.K., Pulping and bleaching of

nonwood fibers, in Pulp and Paper:

Chemis-

try and Chemical Technology, Vol. 1, 3rd

ed., Gasey, J.P., Ed., Wiley, New York,

1980,

pp. 504-568.

21.

Petersen, P.B., Industrial application of

straw, 1991 Wood and Pulping Chemistry,

Tappi Press, Atlanta, pp. 179-183.

22.

Stephenson, J.N., Ed. Preparation of Stock

for Paper Making, Vol. 2, McGraw-Hill,

New York, 1951. Pages 1-91 discuss pulp-

ing of nonwood fibers.

Silica and

kraft

liquor recovery

23.

Grubshein, B.D., Recovery of lime in the

alkaline pulping of straw and reed, Bumazh.

Prom.

36(8):

12-13(1961).

Russian.

24.

Ibrahim, H., Silica is no longer a problem in

the recovery of heat and chemicals from

nonwood plant fibers black liquor, 1988

International

Non-wood Fiber Pulping and

Papermaking

Conference,

Ghina,

877-889.

25.

Lengyel, P., Silica removal from black

liquors, Zellstojfu. Pfl/?/^r 9(3):89-94(1960).

26.

Sawheny, R.S., Desilicanization of black

liquor-a

glance,

1988

International Non-wood

Fiber

Pulping

and

Papermaking

Conference,

Ghina, pp 933-942.

27.

Schwalbe, G.G., Process for desalination of

alkaline silicious spent liquors, German

Patent 522,730 (1929).

HYDRAULICS

30.1 FLOW OF LIQUIDS

Introduction

The material in this section is examined in

more detail in many introductory works such as

McGill (1980). Related concepts have been

presented elsewhere, such as the static head of a

headbox (page 213).

Reynolds number

The flow of liquids in pipes can be difficult

to describe mathematically. In order to simplify

the mathematical development, flow is often

considered in circular—cross—section pipe (as is

done here). The Reynolds number is an important

parameter used to predict the flow pattern in a

pipe.

The Reynolds number is defined as:

Re^

Du.p

where

= Reynolds number

= inside diameter of pipe

= bulk fluid velocity

= fluid viscosity

If the Reynolds number is below 2000, the

flow tends to be laminar or streamline; that is,

the pressure and flow velocity at any point remains

constant with time. Each point on a concentric

circle has the same pressure and flow velocity.

The flow velocity is highest at the center point.

If the Reynolds number is above 2000, the

flow tends to be unstable or

turbulent;

that is, the

flow velocity at a given point is not constant with

time,

although there may be a layer of laminar

flow near the pipe wall. At high Reynolds num-

bers the flow becomes quite turbulent.

The addition of a small amount of fiber to

water (on the order of

0.3%

consistency although

it can be higher at higher flow velocities) may

actually reduce friction compared to water alone.

When this is true, the plug of fibers tends to flow

in the center of the pipe, while a clear water

annulus flows near the edge of the pipe.

Flow

pipes

For laminar flow in smooth—walled, straight,

circular

pipes,

the Hagen—Poiseuille law gives the

flow rate j2 as a function of the pressure gradient

across the length of the pipe (AP), the pipe radius

(r),

the pipe length (L), and the fluid viscosity (jx):

Q =

irAPr^/iSiiL)

The Darcey equation gives frictional losses

(h) as a function of frictional factor (f), pipe

length (L), average velocity (v), and acceleration

due to gravity (g) for flow in circular pipes of

diameter/) as:

h=fLvV(2Dg)

For Reynolds number below 2000, / has been

empirically shown to be equal to 64/Re.

Water hammer

Water hammer is the sudden change in

pressure caused by a column of water that sudden-

ly changes velocity (as when a valve is closed).

30.2 PUMP BASICS

Introduction

Pumps may be considered as positive—dis-

placement or dynamic types. Dynamic pumps

usually rely on centrifugal force. Centrifugal

pumps are the most common type of dynamic

pump and the most common type of pump used in

pulp and paper mills.

Cavitation

and related problems

The inlet of any pump operates at a lower

pressure than the source of the liquid; it is called

the suction side of the pump. This is what causes

the liquid to flow into the pump. If the pressure

at the suction side of the pump drops below the

vapor pressure of the liquid, gas bubbles (cavities)

will form that interfere with the proper operation

of the pump. The gas bubbles collapse on the

output side of the pump where the pressure is

high, making much noise and causing undue wear.

This condition is known as cavitation.

641

642 30. HYDRAULICS

Cavitation is aggravated by using liquids at

high temperatures where their vapor pressure is

high, by using filters with high pressure drops, or

by using inlet lines that have high pressure drops

at the operating flow rates.

Cavitation can be alleviated by having a

generous geodetic height for the liquid entering the

pump; this is accomplished by lowering the pump

or raising the reservoir. Inlet pipes and filters

with larger inside diameters will decrease the

frictional and velocity pressure drops. If the flow

rate is pulsating with a period less than a few

seconds, the use of a dampening system can level

the flow rate at the inlet. In some cases the liquid

in the reservoir can be padded with a pressurized

inert gas, although this can lead to bubble forma-

tion downstream (much like the bends that may

occur with scuba divers). Booster pumps may be

used to solve difficult cavitation problems.

The use of an adequate NPSH will avoid

cavitation. Much difficulty and premature wear

encountered in the use of pumps are the result

of an inadequate NPSH.

NPSH,

NPSHA, NPSHR

NPSH, the net positive suction

head,

is the

absolute suction head minus the vapor pressure of

the liquid being pumped. NPSHA is the net

positive suction head available and must be greater

than NPSHR, the net positive suction head re-

quired. NPSH increases as the square of the

pump flow rate near and above the design flow

rate of a pump. It does not increase this fast at

flow rates much below the rated flow rate.

Good design practices on the inlet piping help

insure an adequate NPSH. Suction piping should

be at least twice the diameter of the pump inlet

nozzle. Pump inlet velocities should be below 10

ft/s. Place pipe elbows at least 5—10 pipe diame-

ters from the pump inlet, with the lower ratio

applicable to large pipe diameters. When this is

impossible, the use of rotating vanes before the

elbow can solve problems. Inlet screening devic-

es,

if used, must not have high pressure drops.

High elevation reduces atmospheric pressure

and NPSHA. Every 1000 feet of elevation reduc-

es NPSHA the equivalent of 1.1 feet of water

head.

Pump selection

One must consider the properties of the fluid

to be controlled (abrasiveness, viscosity, corro-

siveness, and specific gravity), the flow rates in-

volved, the pressure of the system, the tempera-

ture range and gradient to which the pump will be

exposed, and the length and frequency of cycling.

The suction conditions are also very important. If

a pump is being replaced, consider the reason the

old pump is no longer suited to the job.

30.3 POSITIVE DISPLACEMENT PUMPS

Positive displacement pumps actually enclose

the fluid to be moved through the system. The

pressure and volume at which the liquid can be

discharged depend only upon the mechanical

strength and size of the properly operating pump

and the energy supplied to the pump (neglecting

leakage). Pressure relief valves are used to

prevent damage to the system or buildup of high,

dangerous pressures. Priming of positive displace-

ment pumps may or may not be required.

Reciprocating and

diaphragm

pumps

Reciprocating pumps use a piston moving

back and forth in a cylinder (see Fig. 30-16).

Two check valves for the input and output at one

end of the cylinder allow one—way flow; this

gives a pulsed output. Because rotational motions

of engines are not easily converted to reciprocating

motions, this type of pulp is seldom used on

mobile equipment. Two cylinders can be used to

give a continuous output as in pumps for high-

performance liquid chromatography, or one cylin-

der can be made to deliver with either stroke. The

diaphragm pump (Fig. 30-1) is similar in princi-

ple,

except that the piston assembly is replaced by

a diaphragm and housing; diaphragm pumps are

often powered by compressed air and occur as

double—diaphragm pumps.

Rotary

pumps

(gear

and lobe types)

Rotary pumps use a variety of means to

generate flow. A pair of gears with many teeth

(external to the gear and called an external gear

pump) in gear pumps or a few lobes in

lobe

pumps

is demonstrated in Fig. 30-2. Internal gear pumps

use a gear within a gear design; internal means

the teeth of one gear are on the inside, projecting

inward. Gear pumps are useful for pumping

POSITIVE DISPLACEMENT PUMPS 643

Output

Fig. 30-2. Operation of a gear or lobe pump.

Fig. 30-1. Cutaway of a diaphragm pump with

two check valves. Courtesy of Gorman—Rupp.

viscous liquids or slurries, but highly viscous

liquids will require relatively large, slow—moving

pumps to keep NPSHA greater than NPSHR. Heli-

cal screws (like a screw press shown on page 293)

or sets of screws are also used. Piston and pro-

gressing cavity pumps have rotary forms.

A series of vanes in

vane

pumps is shown in

Fig. 30-3. Vane pumps are commonly used on

mobile hydraulic equipment. Fig. 30-3 shows an

unbalanced vane pump; by use of an elliptical-

shaped housing with a central rotor each pair of

vanes can have two inputs alternated by two

outputs for each rotation. Rotary vane pumps are

also used for vacuum pumps.

Progressing cavity

The progressing cavity pump (page 717)

provides axial pumping in a tube—shaped pumping

chamber. The rotor (the inside portion that turns)

is externally threaded or lobed, and the stator (the

outside, stationary portion) is internally threaded

or lobed. Cavities between the rotor and stator

progress from the inlet to the outlet. While rotors

are usually constructed from metal, stators are

often replaceable and made of elastomers.

Peristaltic pump

The peristaltic (or tubing) pump (Fig. 30-4)

uses a tube in the shape of a semicircle with two

or more pinching rollers on the outer edge of a

wheel that rotates on the inside of the semicircle.

These rollers pinch the tubing and enclose small

Fig. 30-3. Operation of a vane pump.

amounts of fluid and push them forward in the

tubing. The pump acts as a check valve. Only

the tubing is contacted by the material to be

pumped so seals are not required. These pumps

are usually fairly small and have flow rates on the

order of less than 1 mL/min at low pressures,

although large units using tubing of

mm inside

diameter can pump 600 L/min at 220 psi. The

pumps are useful for metering materials.

Fig. 30-4. Diagram of a peristaltic pump.