Duan C.G., Karelin V.Y. Abrasive Erosion and Corrosion of Hydraulic machinery

Подождите немного. Документ загружается.

374 Abrasive Erosion and Corrosion of Hydraulic Machinery

surface, however, corrosion damage was mild at the periphery of the spiral

part where the fluid velocity is high, and was severe near the shaft hole where

the velocity is low, and the maximum depth of graphitization have exceeded 2

mm after a 4600-h operation. This may well be attributed to the construction

of a macroscopic corrosion cell where the high velocity zone near the

periphery is cathode and the low velocity zone near the shaft hole is anode.

(a) (b) (c)

(a) Inner view of casing

(b) Magnification of the same as the right

Figure 7.16 Corrosion of seawater pump made of cast iron

(after 4600 hour neat the shaft hole)

Figure 7.17 shows the corrosion current distribution on the inner surface

of a casing (cast iron) which pumped a 3% NaCl solution saturated with

oxygen. Corrosion current was measured by using a zero-shunt ammeter

between a certain piece of fraction (shown by dotted lines) and the rest of the

fractions which were electrically isolated by epoxy resin but connected by

leading wire. The result of this current measurement clearly substantiates the

fact that the low velocity zone near the shaft hole is retained anodic.

The corrosion damage due to the velocity difference can be altered by

changing the material. Figure 7.18 shows the corroded surface of a seawater

Corrosion on Hydraulic Machinery

375

pump casing made of cast iron containing 1% Cr. Compared with Figure

7.16, the degree of the corrosion damage is extremely slight.

(Caused

by velocity

difference)

(Pump

casing made

of cast

iron,

3%NaCl

sol.,

25°C)

Figure 7.17 Distribution of current density of macro-cell corrosion

(a) (b)

(a) Inner view of casing

(b) Magnification of the same as the right

Figure 7.18 Corrosion of seawater pump casing made of 1% Cr cast iron;

Good resistivity compared with the case shown in Figure 7.16

376

Abrasive Erosion and Corrosion

Hydraulic Machinery

The corrosion current near the shaft hole of pump casings was measured

using several materials. Results are shown in Figure 7.19. In the case of plain

cast iron (JIS FC 20), the corrosion current increased rapidly just after the

start up to reach a considerably high value indicating the occurrence of severe

corrosion. The gradual decrease thereafter is apparently due to the formation

of a graphitization layer on the surface which increases its thickness with the

progress of the corrosion. In the case of 1% Cr cast iron (ES

IF, Ebara

Corp.

Standard, low chromium cast iron for seawater pump) and austenitic

cast iron (ASTM A436 type 2), the corrosion current was extremely low

compared with the plain cast iron (JIS FC 20). It is worth noting that the

corrosion resistance of ES 51 F is remarkably improved by adding only

Cr.

1000

£ 500

♦■»

'55

■o

750-

Z 250

-250.

..^

.-FC20

tteetric quantfty

\ ES51F

V^laetrlc

quint

\\436

Type

i i ...

2630 C/cm

,460 C/cm

—-

100 Time (day)

Figure 7.19 Comparison of macro-cell corrosion caused by velocity difference

of several materials (corrosion current density near the shaft hole)

In contrast to cast iron, the high velocity zone of a tin bronze (FIS BC6)

casing became anodic. This may be attributed to the fact that the corrosion

product layer on die tin bronze surface is easily removed by a fluid flow of

high velocity and dissolution of substrate is enhanced there. This difference in

the influence of flow velocity between iron based alloy and copper based alloy

is in accordance with the experimental results obtained by Laque in 1972

using the rotating disk test apparatus.

Corrosion on Hydraulic Machinery 377

7.3.2 Corrosion Promoted by Mixed Use of Different Materials

(Galvanic Corrosion)

Initially, use of stainless steel in large-sized seawater pumps was confined to

some limited parts only. The remaining parts of the pump were of plain cast

iron which acted as a sacrificial anode to protect the stainless steel parts from

corrosion. This led people to the unfortunate misunderstanding that stainless

steel was a perfect material for use in seawater, which further accelerated the

use of stainless steel in seawater pumps. The increase of the surface area of

stainless steel (cathode) and the decrease of the surface area of cast iron

(anode) increased the rate of corrosion on the cast iron parts, which made cast

iron lose its reputation as a construction material. Pump designers and users

made another mistake in introducing tar-epoxy coating in order to reduce the

corrosion damage. If they had had right knowledge of corrosion theory, they

would have coated the stainless steel parts in order to reduce the cathode area.

They chose, however, to do the opposite; the cast iron was coated and the

stainless steel remained uncoated. What was worse, those portions of

the

cast

iron parts which were in direct contact with stainless steel, e.g., fringe surface

or faucet joint surfaces, remained uncoated in order to maintain precise

measurements. As a result, corrosion currents concentrated on the uncoated

surface causing decisive damage there in quite a short period (Figure 7.20).

Pinholes on the coating layer also brought about severe damage (Figure 7.21).

Recently, people have gradually obtained a correct understanding of the

problem and have begun to devise adequate means for its prevention. Figure

7.22 shows an example of such countermeasures; an isolating phenol resin

plate as well as isolating bolts and nuts used between the stainless steel parts

and plain cast iron parts.

The possibility of galvanic corrosion caused by the coupling of materials

other than stainless steel and cast iron has of course to be examined before

any combination is put into use. Various kinds of seawater corrosion tests

were conducted from 1977 for five years at several power plants located

along the coast Japan. Figure 7.23 shows the assembly and test piece for one

of the above mentioned galvanic corrosion tests; in test piece A, the

anode/cathode ratio was 1/1, and was variable in test piece B through the

regulating of

angle

378

Abrasive Erosion and Corrosion of Hydraulic Machinery

Figure 7.20 Galvanic corrosion of case iron part conducted

to stainless steel part in seawater. Flange of the part (25 mm thickness)

was perfectly graphitized only 2-year operation

Cavity by graphitize

corrosion under defects

of coatings

Figure 7.21 Localized galvanic corrosion at the pinhole of coating

on cast iron part conducted to stainless steel part in seawater

Graphitization occurred concentrically to 20 mm depth at this point

only 2-year operation

Corrosion on Hydraulic Machinery

379

Figure 7.22 Application of an isolating plate and isolating volts

seawater pump

lest piece SUS304

nut for

sacrificial •

anode(SS4l2

support Bolt

(SUS304)

nut for f ixing(SUS304)

^washer (sussw)

isolating washer

v(PTFE)

(phenol resin) /

distance piece

(PTFE)

test piece

'4-M8 (polycarbonate)

Test piece

(anode: cnthode=l

: I)

•poxy resin

Assembled view

test pieces

SUS3<xU/L

test piece

Test piece

(variable anode/cathode ratio)

Figure 7.23 Assembly and test piece for galvanic corrosion test

380

Abrasive Erosion and Corrosion of Hydraulic Machinery

In the case of 1/1 ratio coupling of tin bronze (JIS BC 6) and stainless

steel (JIS SUS 304), the former was anodic and its corrosion rate was

increased to 1-6 times that of it's isolated condition. The corrosion rate ranged

from 0.07 to 0.45 mm/year. In the coupling of aluminum bronze (JIS A1BC 2

or A1BC3) and stainless steel (JIS SUS 304), the bronze was anodic and its

corrosion rate was increased to 1 -3.7 times that of it's isolated condition.

The corrosion rate was 0.02 ~ 0.2 mm/year. A decrease in anode/cathode

ratio brought about some crevice corrosion on the stainless at contact surface.

By the coupling of austenitic cast iron and plain cast iron, the austenitic

cast iron acted as cathode to promote corrosion in the plain cast iron (see the

first column of Table 7.2). On the other hand, the austenitic cast iron was

seriously damaged when it was coupled with stainless steel as shown in the

last column of Table 7.2.

Table 7.2 Galvanic corrosion in natural seawater at several points

throughout Japan (Anode/Cathode =1)

Geographical

location

Tomakomai

Sendai

Fukushima

Kashima

Chiba-1

Chiba-2

Naooya

Hlmeji

Fukuyama

Tokuyama

Kokura

5 ita-1

0 ita-2

Naoasaki

FC20/A436 type2

0.87 mm/Y

0.65

0.38

0.45

0.83

0.29

0.38

0.44

1.37

0.42

0.46

1.15

0.74

Galvanic

couples

FC20/SUS304

1.12 mm/Y

0.76

0.49

0.56

0.79

0.80

0.29

0.38

0.41

1.43

0.41

0.43

1.18

0.99

A436 type2/SUS304

0.95 mm/Y

0.50

0.35

0.48

0.49

0.17

0.19

0.18

0.63

0.21

0.14

0.39

0.27

Polarization curves for these materials in a

NaCl solution are given in

Figure 7.24 which indicates that the austenitic cast iron acts as cathode to

Corrosion on Hydraulic Machinery

381

promote corrosion of the plain cast iron, but it dissolves itself as anode when

it is coupled with the stainless steel. Thus the polarization curves substantiate

the results of Table 7.2.

-1.0 -0.8 -0.6 -0.4 -0.2

Potential ( V

vsSCE

)

Figure 7.24 Polarization behavior of gray cast iron, austenitic

cast iron and stain-less steel (3%NaCl, 25°C)

7.3.3 Crevice Corrosion and Other Localized Corrosion

(I) Crevice corrosion

Large-sized and yet full stainless steel pumps come into general use in

accordance with the rapid economic development in Japan. They have,

however, a weak point as regards crevice corrosion. Countermeasures to this

type of corrosion are:

i) To use sealing material to prevent the permeation of seawater into

the crevice;

ii) To use crevice corrosion resistant materials.

The first measure is not satisfactory for long term use and can not be

applied to the crevices on moving parts. As for the second measure, the new

types of stainless steels of higher Cr and Mo contents have been utilized and

proved considerably effective in preventing crevice corrosion. Even these

materials, however, are not totally immune from crevice corrosion.

Table 7.3 is a result obtained from crevice corrosion tests which were

conducted parallel with the galvanic corrosion tests described in the

382

Abrasive Erosion and Corrosion of Hydraulic Machinery

proceeding paragraph, using the assembly shown in Figure 7.25. High graded

materials such as Alloy C, Alloy 625 or Tiganium are certainly reliable, but it

is economically impossible to use such materials for large-sized pumps.

Kitashima et. al. found the way out of

the

problem by developing an alloy for

welding use which has superior resistance to crevice corrosion and by limiting

its application to crevice portion only. Figure 7.26 shows an example of

successful application to the overlay welding.

Table 7.3 Crevice corrosion in natural seawater at several points

throughout Japan.

Geographical

location

Tomakomai

Chiba-1

Nagoya

Tokuyama

Oita-1

sus

304

X X

Alloy Alloy

825

X X

Alloy Alloy

Ti Duplex

625

o: generally free from crevice attack

A: occasionally subject to crevice attack

x: generally subject to crevice attack

x: generally subject to severe crevice

attack

test piece

distance piece

asbestos

sheet packing

PTFE tube

support

Bolt

Figure 7.25 Schematic of test assembly for crevice corrosion test

Corrosion on Hydraulic Machinery 383

(a) Flange surface of column pipe (base material; SUS316L)

(b) Flange surface of intermediate bearing support (base material; SCS 14

solution heat treatment after overlay).

Figure 7.26 Parts of seawater pump weld overlaid with corrosion resistant

alloy (Ni 30%

Cr-10%

Mo alloy) after

year operation.

(2) Corrosion of weld metal:

A similar material to the base metal is usually used for plugging welding of

the vent hole or for repairing welding of the defects in castings. Unexpected

damage can occur even in these weldings. Figure 7.27 (a) shows an example

of severe corrosion damage to the deposit metal at the plugging welding. The

base metal was type 304-cast stainless steel (JIS SCS 13) and the

environment was a dense saline solution of 70 ~ 90°Ccontaining solid

particles. Investigations revealed that the microstructure of the deposit metal

was normal and there was no sign of sensitization as shown in Figure 7.27

(b).

A little difference, however, was found in chemical composition between

the base metal and the deposit metal; deference

Mo content should be noted

in Table 7.4 which shows the chemical composition of the metals, i.e., the Mo

content of the deposit metal (0.006%) is lower than that of the base metal

(0.67%).

The contents of other elements were quite normal for type 304

stainless steel, specifically; levels of Ni and Cr were rather higher than in the

base metal.

Test specimens with an exposure surface of 2 cm

were prepared from

each metal and dipped into a pseudo-environmental liquid (without solid

particles) in order to measure the potential as well as the galvanic current