nuclear pressurized water reactor (pwr) - mechannical design materials

8/17/2019 Nuclear Pressurized Water Reactor (PWR) - Mechannical design materials

1/95

XA9949546

3. REACTOR PRESSURE VESS EL MATERIALS

K. Suzuki, The Japan St eel

Works,

Ltd., Muroran Plant

1 INTRODUCTION

The demands placed on rea ctor pressure vessel (RPV) steels are

severe. They must be manufactured in required sizes and

thicknesses, be of sufficie nt strength and toughnes s, show little

deterioration under irradiation, allow the production of high

quality welds and be compati ble with the cladding. This Chapter

refers to non-WER press ure vess els, but many of the guiding

principles described here apply to that particular case.

Starting with carbon steel plates and forgings for conventional

boiler drums in the dawn of commercial light water reactors (LWRs)

followed by a few change s thereafter, SA533 and SA508 and similar

grade steels have become well established [1 ]. Both grades are of

the vacuum treated, que nche d and tempered type of 600 N/mm

2

strength

class, which is not the highest level in weldable structural steels.


2/95

- 71 -

The specific requirements for RPV steels are to giv e, even to l arge-

size component materials of

R P V s ,

higher values of the following

properties:

- uniformity and isotropy of mechanical properties, including

less mass effect in the mid-section

- fracture toughness

- internal defects

- weldability

- resistivity to neutron irradiation embrittlement

Some further items were added during the past two decades primar ily

for the purpose of easier execution of non-destructive examinations

both pre-service and in-service. These are :

- fewer weld seams in RPV

- larger and more integral design of component materials

The requirements have been steadily realized.


3/95

- 72 -

With respect to the weld, improvements have been made by adequate

selection of flux in combination with requirements on the chemical

composition of the weld wir e, welding parameters such as weld bead

size and seque nce to minimize the coarse grained areas in the heat

affected zone (HAZ) [2 ].

2 HISTORICAL REVIEW [3]

Many of t he earlier plants were constructed of so-called carbon or

mild steels (usually in normalized and tempered, NT, c o n d i t i o n ) , but

several of these were prototypical units and applications have moved

to the wid ely accepted low alloy mangan ese, molybdenum, nickel,

quenched and tempered (QT) grades of higher strength. Early gas

cooled reac tors were also of carbon steel, but these were replaced

in later vers ions by prestressed concrete vessels.

Most of th e reacto rs operating at mid-197 3 were constructed of the

manganese-mo lybdenum steels in QT conditio n. In practice, the use

of the low alloy steels has predominated. For example, all of the

Japanese react ors o f BWR and PWR types have A533-B Class 1 (QT)

pressure ve ss els except Tsur uga which uses A302-B (QT) and JPDR-2

with A302-B in NT condition [4] . The sole Japanese gas cooled

reactor (Tokai Nuclear Power Sta tion) was contained in a vessel of a

Japanese C-Mn stee l. Typica l composition data for the Japan Steel

Works

( J S W ) ,

Ltd. steel use d is C-0.10*, Mn-1.30%, Si-0.25%, P-


4/95

- 73 -

0.014%, and S-0.0 18* (equivalent to the UK steel, BS NDIV originally

planned for this reactor

v e s s e l ) .

Special care was taken to

minimi ze p hosp horus and sulphur and to refine grain size, thereby

enhancing no tch toughness. In addition, an experimental programme

was conduc ted to develop the best welding electrode material for

joining plat es of the Tokai vessel [ 5] . The relative content of

mangane se and silicon was determined for optimum toughness of welded

structures.

The Ag esta reactor of Sweden was contained in a carbon steel,

equivale nt to ASTM Type A212-B (a carbon-silicon-manganese

s t e e l ) ,

the Osk ars hamn- 1 vessel s teel was equivalent to ASTM Type A302-B [6]

(a manganese-mo lybdenum

s t e e l ) .

The A212-B steel was also used in

on e of t he e arl y U.S. reacto rs, Indian Point-1, and a limited number

of experimental reactors.

In the Fede ra l Republic o f Germany (F.R.G.) the steel designations

matc h nati ona l conventions, but most water reactor steels were quite

similar to U .S . grades A508-C1.1 for forging and A533-B for plate

steel [7 ]. And the most RPVs were strongly dependent on the 22

NiMoCr 37 composition (similar to A 5 0 8 , C1.2 : a nickel-chromium-

molybdenum

s t e e l ) .

22 NiMoCr 37 steel was used in F.R.G. until 1976

and fulf illed all requirements [8 ]. However, this type of steel

exhibited so me susceptibility to stress relief cracking and

underclad cracking [ 9 - 1 1 ] . By this reason the use of 20 MnMoNi 55

steel (similar to A 5 0 8 , C 1.3 : a manganese-molybdenum-nickel steel)


5/95

- 74 -

was increased.

Little basic difference in steel type exists among the vesse ls of

reactors now operating throughout the world. Howev er, certain

metallurgical differences of crucial importance are identificable,

but other considerations such as neutron flux and fluence,

irradiation temperature, stress state etc. have equa l o r gre ater

bearing on radiation embrittlement sensitivity. Relati ve to

radiation embrittlement, however, it is important to realize t hat

a l l s t e el s u s e d t o d a te h a v e t h e s am e b a s ic w e a k n e s s . T h e y a r e

susceptible to radiation hardening, increases in stre ngth, and

reduction to fracture toughness. Thus, it is appropriate, eve n

necessary, to generalize initially in describing radi ation

embrittlement of vessel steels.

The curr ent choice of material is of specific A-53 3B, Cl.l allo ys

f o r p l a te s a n d A-508, C1.3 alloys for forgings (Table 1 ) [1 2] .

These ar e ve ry similar steels of the low alloy m anganese-mo lybdenum

type (A302-B) with nickel added. Significant differences have be en

developed between the old and new steels. Major differences ar e in

improved strength and toughness for the new s teels with a change of

micro -struc ture from basically pearlitic to tempered bainitic t o

temperecd martensiti c or mixed bainitic-martensitic microstr uctur e.

Chemically, the new steel is purer, that is, it contains smaller

amounts of r esidual or tramp elements, because steel-making

procedures incorporating vacuum degassing steps are applied

routinely in their production. These steps , producing singifican t


6/95

- 75 -

changes,

are considered improvements from the vi ew o f r adiati on

embrittlem ent, but the resulting implication is that the gr eates t

concern must be addressed to the older plants , where our knowled ge

of the alloys used and the service conditions is least. Know ledge

about welds is probably both the most critical a nd the most limited.

Emphasis then must be placed on learning mor e about the old systems

material s, especially welds , and improving the new (future)

materials.

3 BASE MATERIAL

3.1 Chemi cal Composition [12]

Specifications for the structural steels whic h ar e extensiv ely used

for the LW R components in Germany and in the Unite d States ar e

listed in Table 1 [12].

Som e basic investigations were conducted to confir m th e effect of

chemical composition on 20 MnMoNI 55 in addition to the basic

studies for heavy forgings [10, 1 3 ] .

Fig.l s hows the continuous cooling transfor mation curves (CCT curve)

for this steel with 0.17% and 0.20* C. The differences in ferr ite,

pearlite and bainite transformation range ar e obser ved due to a

little difference in C content. Also the effect of M n, Ni, Mo on

the hardenability was investigated, but the res ults d o not indica te

substantial differences in the CCT curve.


7/95

- 76 -

In order to main tai n sufficient toughness o f large forgings at lower

temperature , it is important to provide fine gr ain as well as proper

structure.

There a re some procedures to refine the grain size, and

it is an usual procedure to add Al. During the quenching of large

and mass ive for ging , austenitizing requir es considerably long time

to obtain uniform heating. In this case, the presence of AIH is

strongly effective in preventing grain growth.

V is know n as a predominant element to enhance the tensile

properties of material. However, the toughness and weld crack

sensitivity o f the material are also affected by the addition of

this elemen t. Thus no V should be added, even though the maximum

content of 0.03 or 0.05% is allowed in the mater ial specifications-

It is said th at Cu and P affect the irradiation damage which is

evaluated by the shift in Charpy impact curve [1 4] . In response to

this requir ement, the target values of Cu and P content of 0.0835

max.

and 0.0 08% max . respectively are maintained.

Concerning weldabi lity of the steel, liquation cracking and stress

relief cra cking in particular, crack promoting elements and

threshold value s for cracking occurrence were investigated.

Cracking ca n be avoided b y limiting the Mo cont ent, the elements P,

S, Cu, Sn, N, As , Co and Al as well [11] .


8/95

- 77 -

3.2 Steel and Ingot Making [15]

3.2.1 Manufacturing Process of Large Ingots

(1) History o f the production of large ingots

In F i g . 2 , majo r changes in the history of the production of large

ingots in JSW are summarized. It also shows the transition of

maximum ingot size. Before Bochumer Verein-type vacuum casting

facilities wer e installed in 1959, steel was melted by acid open

hearth fu rnaces to minimize hydrogen pick up and the steel cast in

air.

After t he installation, acid open hearth was replaced by basic

open hearth and electric arc furnace ( E A F ) , because hydrogen removal

was made pos sibl e during vacuum casting. In order to obtain higher

degree of vac uum during casting, a steam ejector was introduced in

1 9 7 0 . In 197 3 a holding furnace was installed to replace the open

hearth. Accor ding to the demand from industries, sizes of ingot

became lar ger year after ye ar . In 1969 a world largest 400 ton

ingot was pr oduced, and the record was soon renewed by 500 ton ingot

in 1971, 570 ton ingot in 1980 , then 600 ton ingot in 1985.

(2) Installation of ladle refining furnace (LRF)

In 1980, to m ee t higher requirements for the record 570 ton ingots

of nuclear ve ss el application, vacuum facilities were installed to

the holding furnace to convert it into a ladle refining furnace.


9/95

- 78 -

Fig.3 shows a schematic outline of the furnace. It has on e heatin g

system and two vacuum systems so that two vessels o f mo lten steel

can be treated at a time. Combining vacuum treatment at the LRF

with conventional tap degassing the "double degassing" pr ocess was

developed for the production of large ingots up to 600 ton. Fig .4

shows the production sequence of a 600 ton ingot. By this pro cess

the quality of products such as for nuclear applicatio ns wa s

remarkably improved.

(3) Gene ral description of large forging ingot

Fig. 5 shows a sulphur print of the longitudinal secti on of a 75 to n

ingot (left) and the description of segregations and solidif ication

structures o f the ingot

( r i g h t ) .

In a sulphur print solute-enriched

portions ar e marked dark. In the region called "branched column ar

zone", string-shaped A-segregates are observed and in the re gion

called "equiaxed

zone",

V-segregates are observed. In these two

regions, microporosities are easily formed. Non-metallic

inclusions , especially oxides, are often observed in the re gion

called "sedimentation

zone"

at the bottom of an ingot and thi s m ay

lead to the rejection of the entire ingot. Distribution o f C and of

other alloyi ng elements is not uniform. Generally speaking, thes e

elements a re rich in the top side of an ingot and poor in the

sedimentatio n zone. In the following section, technical

improvements to reduce these heterogeneities will be descr ibed.


10/95

- 79 -

3.2.2 Rec ent Improvement in Steelmaking and Ing otmak ing

(1) Removal of P and S

Low S and P contents about 0.004-0.008% can be obta ined by

conventional EAF process, however, LRF process wa s introduced to

obtain extr emely low S and P contents in the liquid s t e e l . F i g . 6

shows a flow chart of the refining process by LRF. A high degr ee of

desulphurization up to 90 % can be obtained by using high bas ic slag

and extreme ly low content of S less than lOppm can be obtai ned. Th e

lowest recorde d S content is 2 ppm. Since P cannot be remove d by

LRF tre atment, it is essential to cut off oxidizing slag fro m EAF to

prevent rephosphorization. Fig.7 shows the changes of P content an d

S content whe n extremely low content of P+S


11/95

- 30 -

(3) Degassing (H, 0) and reduction of non-metallic inclusions

(a) Effect of LRF

Degassing is achieved ver y effectively by LRF process due to high

vacuum and intensive stirring of molten steel. For the stirring, Ar

gas is blown into molten steel through porous plug at the bottom of

refining ladle. Energy dens ity of Ar stirring *., (Watt/ton) is

expressed by :

r

M

= 6

'

1 8 Q T

L (ln(l+

H

) + ( -

T

o ) }

148Pa T

L

Where

Q is flow rate of Ar-gas (Nm

3

/ m i n ) ,

Pa is pressure at the surface of molten steel ( a t m ) ,

T^ is temperature of mol ten steel (K) ,

T is temperature of Ar gas before blowing (K ),

M, is weight of molten steel ( t o n ) ,

H is bath depth (cm) ,

and t is treatment time

( s e c ) .

Fig.8 shows th e effect of

g

M

»t

on the degree of hydrogen removal.

In this process a high deg ree of hydrogen removal, up to 80% , is

obtained by increasing the "stirring" intensity e

M

»t , Fig.9 shows

a change in hydrogen content during LRF process . When molten steel

is cast by botto m pouring, about

0.5ppm

of hydrogen pick-up should


12/95

- 81 -

be ex pected, whereas if it is cast by mold stream degassing, low

hydrogen levels of 0.4 to

0.6ppm

are obtained- Fig.10 shows the

e f f e c t o f

f

M

» t

o n t h e d e g r e e of d e o x i d at i o n . A t t h e r eg i o n o f l o w

£

M

«.t

degree of dexidation increases with increasing jw»t

however , it decreases with further increase of

£

M

»t »

It is

a t t r i bu t e d t o t h e e r o s i o n o f b r i ck o r s u s p e n s io n o f s l a g i n t h e

m e l t .

By controlling the stirring intensity g

M

«t a r o u nd 8 0 - 1 0 0 x l 0

3

J/ton, low oxygen level of less than 20ppm is obtained.

( b) E f f e c t o f do u b l e d e g a s s i n g

A s d e s c r i b e d b e f o r e , t h e d o u b l e d eg a s s i ng p r o c e s s ( va cu um t r e a t m e n t

a t L H F p l u s m o l d s t r e a m d e g a s s i ng ) i s v e r y e f f e c t i v e f o r d e g a s s i n g

o f m o l t e n s t e e l a n d w a s a p p li e d t o t h e p r o d u c t i o n o f t w o 5 7 0 t o n

i n g o t s o f n u c l e a r a p p l i c a t i o n s . F i g . 1 1 s h o w s t h e c h e ck a n a l ys e s o f

h y d r o g e n a n d o x y g e n o f t h es e i n g o t s . L o w h y d r o g e n c o n te n t s o f l e s s

t h a n l p p m a n d o x y g e n c o n t e n t s a r o u nd 1 0 t o 2 0 p p m a r e o b t ai n e d i n t h e

b o d y o f i n g o t s . F r o m t h i s l o w g a s c on t e nt t h e e f f e ct o f d o u b l e

d e g a s s i n g i s e v i d e n t .

( 4) R e d u c t i o n o f m a c r o s e g r e ga t e s

F o r 0 . 7 * c a r b o n s t e e l , c r i t i ca l c o n d it i o n f o r t h e f o r m at i o n o f A -

s e g r e g a te s i s e x p r e s s e d b y :


13/95

- 82 -

Where

j is cooling rate in the radial direction of an ingot (°c/min)

and R is solidificatin r ate in the radial direction of an ingot

( m m / m i n ) .

By changing the constant term in the right side, the equation i s

applicable to other kinds of steel. If carbon content and silico n

content ar e hig h, the constant is large and A-segregates are easily-

formed. Mo and Cr have the opposite effect. In general it is

difficult t o control j and R during the solidification of an actual

ingot,

theref ore, chemical compositions are adjusted to minimize the

formation of A-segregates. Since the driving force for the

formation o f A-segregates is considered to be the density diff eren ce

between solute-enriched liquid and bulk liquid, the density

differnece ApL

i s

calculated assuming solute enriched liquid is

that at fra ction solid 0. 3* ApL varies with various steels and if

is large A-segregates are easily formed. The relation between

1 1 a n d A pL

i s

shown in

F i g . 1 2 .

If type of steel is deter mined ,

is calc ula ted and critical value j.R

1

- * is obtained from

Fig.12 and then the position where A-segregates first starts to fo rm

is estimated.

V-segr egates are formed due to rapid shrinking along the axis of an

ingot at the final stage of solidification. V-segregaties can be

reduced by ma king the taper of ingot larger and height-to-diameter

ratio H/D sm aller.


14/95

- 83 -

(5) Redu ctio n of microporosity

Micropo rositi es are formed in the A-segregation zone and V -

segregation zone. If sizes of pores are large, it may result in the

rejection of the product. Therefore, it is very important to des ign

the optimum shape of the ingot to reduce porosity. Based up on

investigations on 3 - 220 ton ingots of carbon steel and low-all oy

steel,

critica l conditions for the formation of microporosity wer e

determined as shown in Fig.

1 3 .

Large porosity tends to form

especially in the V-segregation zone.

The length of the porosity zone Vy (mm) is expressed as a funct ion

of height-to-diameter ratio H/D as shown in Fig. 1 4 . In the figure,

total weigh t of ingot is kept constant and H/D is changed and it is

kno wn that Vy decreases with decreasing H/D. Porosity along t he

axi s of ing ot can be reduced also by increasing the weigh t o f th e ...

feeder hea d. In this case €he rate of axial solidification beco mes

slower and the formation of large porosity are suppressed. Bas ed

up on the abo ve mentioned studies, optimum ingot shapes were d esig ned

and satisfact ory results obtained.

(6) Reduc tion of carbon segregation

Quantita tive understanding of carbon segregation of ingots is ver y

important. However, studies to date have yielded improve ments,

because c ar bo n segregation is affected by many complicated fact ors .

The e quation given below shows the result of multiple regres sion


15/95

- 84 -

analysis on 75 to 570 ton Mn-Mo-Ni steel ingots (number of ingot :

8 1 ) .

Segregation Rate [ - ( S Q - Q K ) adle

x 1 Q 0 ]

(*C)ladle

= 38.5+0.132(Wt)-72.8(F.H.R.)-100(/\C multiple)

Where

(%C) is check carbo n analysis i n t h e centre of (ingot

body/feeder head) boundary face

( w t & ) ,

(%C)

l a d l e i s

ladle analys is

o f

carbon content

( w t % ) ,

(Wt) is total ingot weight ( t o n ) ,

(F.H.R.)

is

feeder head ratio defined

a s

(weight

o f

feeder

head)

/

(weight

of

ingot body )

and (̂ C multiple) is carbon content differenc e at multi-pouring

practice defined as

£

{| (%C)i-(*C)

ladle |

x W i }

i - 1

i - 1

where W i i s t h e weight and (9SC)i is the carbon content o f molten

steel cast as "i"th time.

From the equation above, it is known that large feeder head ratio

and large carbo n content differences favour

t h e

minimizing

of

carbon

segregation

o f

ingots. Fig. 15 shows

t h e

distribution

of

carbon

content alon g the axis o f a 1 4 0 t o n a n d a 1 8 0 t o n rotor ingot cast

by multi-pouring method.

It is shown that these distributions a r e uniform compared with

ingots cast b y conventional method s. Fig.16 shows the distrubution


16/95

- 85 -

of carbon content in the longitudinal section of 570 ton ingot for

nuclear application. The carbon segregation is not as severe as in

a "huge" ingot . It is considered that this results from the pouring

process.

If this ingot had been poured without a carbon content

difference, the carbon segr egation would have been about 109s

( 0 . 0 2 9 J C )

higher according to the estimation by the equation given

above.


17/95

- 86 -

3.3 Forging Process [16]

F i g s . 17 and 18 show the combined vessel flange and nozzle belt

forging of KWU/1300MWe pressurized water reactor pressure vesse l

(PWRPV) made from a 400 t on ingot and the combined vessel flange and

nozzl e belt forging of WEC/1 57" PWRPV made from a 500 ton ingot

(developed by COCKERILL) as compared with the conventional one,

respective ly. These integrated flange forgings were hot worked by a

10,000 ton forging pre ss. Fig. 19 illustrates the forging proce sses

for the flange forging ma de from a 500 ton ingot. Sufficient

discard was made from each end of the ingot to insure that only

sound metal enters the completed forging. After piercing of ingot

c o r e , repeti tion of enlarg ing and upsetting was performed to close

possible porosity inside the ingot.

Th e for ging ra tio and repetition of enlarging and upsetting

operatio ns play an important role to improve the mechanical

prop erti es. Fig.20 shows the improved impact value due to

repetit ion of forging. Theref ore, from the results a minimum

forging r atio of 1.5 s hould b e required.

Anisotr opy o f forged materi al was intensively investigated by W.

Coupette in 1940th [17 ]. Fig.21 shows the effect of forging ratio

and anisotr opy of 20 MnMo Ni 55 steel. Compared with Coupette's

dat a, anis otro py of forging at present time seems to be much

smaller.


18/95

- 87 -

To es timate the mechanical properties due to differences in forging

ratio, the use of the logatithmic strain concept is convenient. The

logarithmic strains in three directions are defined in

Fig.22.

From the results of mechanical tests for different components, the

relation betwe en logarithmic strain and mechanical pro perties, such

as tensile properties and Charpy V-notch impact value , can be

obtained, as shown in

Fig.23.

Tensile strength, yield strength and

elongation, at both room temperature and 350°c, are not so

significantly related to logarithmic strain, but the reduction of

area is increased with the increase of logarithmic strain.

Charpy V-notch impact values are also increased with the increase of

logarithmic stra in. The integrated flange forgings made from a 400

ton and a 50 0 to n ingot respectively were manufactured under the

above consideration. The logarithmic strains of integrated flange

forgings were as given below:

a) Shell flange :

£

t = Q . 9 6

£ a

= -0.28

fit - -0.68

b) Mono-blo ck vessel flange : £ t = n ni

ea = -0.18

fi

r

= -0.63

Therefo re, anisotropy in three directions is expected to be

minimized.


19/95

- 88 -

3.4 Heat Treatment [16]

Heat treatment technology for large forgings is based on the

experimental investigations and experiences of actual ope ratio n in

addition to heat treatment theory. The important points to be

considered are t he segregation in large ingots, mas s effect s,

hydrogen induced defects, temper embrittlement and residual

stresses.

Flake,

ghos t crack and fish eye etc. are well known hydrogen induced

defects. Recentl y, the problems in large forgings due to hydr oge n

are remarkab ly decreased by the improvement of vaccum treat ment

technology.

However, it is still necessary to consider the prevention of

hydrogen induced defects depending on the size and grad e of

materials.

In general, the measures taken for the preven tion o f

hydrogen induced defects are as follows.

- Slow cooling after forging or rolling

- Isoth ermal annealing (Pearlite transf ormat ion)

- Normali zing and tempering (Bainite trans formation)

For large pie ces o f low alloy steel, normalizing and tem pering

technique is usuall y applied. Preliminary heat treatment after

forging sh own in Fig.24 is one example adopted in Germany for

preventing hydrogen flake [ 18] . In this procedure, the bainite

transfor mation is mainly performed after hot working of the fo rgi ng.

At pre sent, in JSH, the basic idea behind this heat tr eatment

diagram is als o being adopted for the preliminary heat tr eatme nt o f

large forgings such as integrated


20/95

- 89 -

flange forging made from 20 MnMoNi 55 steel under the consider ation

of CCT curve shown in Fig.l.

After preliminary heat treatment (normalizing and t e m p e r i n g ) , the

forgings a re machined to a simple cross-section for perfo rming

ultrasonic examination. After successful examination, the forgings

are contour machined further for quenching and tempering in order to

obta in a go od quenching effect. The austenitizing temperatu re of

870 to 910 °c is selected to minimize the grain growth and a faster

quenching operation is for a more complete transformation results .

3.5 Prop erties of Integrated Flange Forgings [16]

3.5.1 Metall urgic al Homogeneity

Segregati on is of vital importance with respect to weldin g and

neutr on irradiation. This is particularly

true,

if the carbon

cont ent is bel ow 0.179s, the mechanical strength re guir ement s a re no t

satisfied, an d if the carbon content is higher than 0.2 4%, th e

weldabilit y is deteriorated due to hardening of the heat affected

zone ( H A Z ) .


21/95


22/95

- 91 -

of mechanical properties is brought about by the homogeneity of

chemical composition and a n accurate and narrow range temperature

control at quenching and tempering.

(3) Directionality of mechanical properties

Fig.28 shows the directionality of tensile properties in three

directions (tangential, axial and radial) for KWU's shell flange.

There is no difference in both yield and tensile strength.

Slight differences in elonga tion and reduction of area are found,

but the difference is not remarkable.

Fig.29 shows the Charpy V-notch impact properties transition curves

for tangential, axial and r adial directions for KWU's shell flange

and COCKERILL's mono-blo ck vessel flange. Impact energy in all

three directions is high e noug h, and any directionality o f

properties is not significant.

The homogeneity in mecha nical properties described above is brought

about by decreasing the micro-segregation and non-metallic

inclusions. Thi s was accomplished by the deeper understanding of

steel making a nd using sui table forging techniques.


23/95

- 92 -

4 WELD

4.1 Structu ral Weld [2]

4.1.1 Wel d Metal

(1) Requirement

Since the plat es and forgings must be welded together, it is obvio us

that the mechanical p roperty requirements of the welded region a nd

the asso ciated heat affected zone (HAZ) can be no less demanding

than those o f the base material itself, particularly as experience

shows that the most likely location for flaws is in the weld and

H A Z .

The ASME requirements for weld mechanical properties and

procedures are somewhat dispersed but appear mainly in Sections I , I

(NB-2300 an d

N B - 2 4 0 0 ) ,

and IX of the Code. They appear to be less

specific t han those for plates and forgings but there is a

requirement that all weldments should conform to all of the mini mum

mechanical proper ty specifications for the materials which are

joined by

welds.

This is clearly a desirable requirement.

Procedures and requirements for Charpy impact tests in particular

are given mainly in the ASME Code Section I . Minimum tensile and

notch toughne ss requirements which are specified in the ASME Co des

are summarized in Table 2. Their extension to cover the

deteri orati on o f properties under irradiation is implemented in 10

CFR 50 Append ix G and in the ASME


24/95

- 93 -

Code Secti on I Appendix 6. Section I (Part C) of the ASME Code

includes general specifications for welding materials and methods

while Se ctio n IX deals with qualifying standards for welding

procedures as well as with the qualifications of the welders

themselves.

(2) Chemical composition

Deposit compositions of manual metal arc welds, associated with

nuclear ves sel s fabricated in Europe and the USA (Tables 3 and 4 ) ,

show that consumables have been employed which are capable of

alloying the deposit with Mh-Mo, Mn-Ni-Mo and Mn-Ni-Cr-Mo. Depo sit

strengths in the post-weld heat treated condition matching tha t o f

the base steel can be achieved by various combinations an d levels of

the elemen ts carbon, manganese, nickel, chromium and molybdenum, and.

this explai ns the variety of deposit analyses that can be foun d in

the litera ture. A similar situation is evident for submerged arc

welds where wires alloyed with Mn-Mo, Mn-Ni-Mo or Mn-Ni-Cr-Mo have

been used b y fabricators to achieve the required deposit strengt hs

(Tables 3 to 5 ) . However, the choice of flux is a very impo rtant

factor go ver ning the deposit composition of submerged ar c w e l d s ,

influencing in particular the carbon, silicon and manganese levels

and the im puri ty element levels of sulphur, phosphorus and o xyg en.

Fused f luxe s o f the calcium silicate type are favoured by m any

nuclear ves sel fabricators. These tend to give lower carbon levels

but higher silic on and oxygen levels in the deposit compared wit h


25/95

agglomerated fluxes [22] , which are chemically mor e basic and

favoured by some European fabricator s, because the ass ociated

deposits generally have higher toughness . Most flux types will add

small amounts of phosphorus but the basi c fluxes a re capable of

lowering deposit sulphur levels, unlike the fused calicium si licate

fluxes.

For the beltline regions, where it is necessary to limit the copper

content to reduce the sensitivity to irradiatio n embrittlement, it

is necessary to depart from the practice of using copper-coated

electrodes, Hawthorne [21] has shown that copper contents belo w 0.1

wt9» are readily achieved i n sound wel ds providing the weldin g wi res

are protected from corrosion before use.

(3) Welding procedure

The main aims of evolving a satisfactory welding pro cedure ar e to

obtain the required mechanical properties in the weld, namely

strength and toughness, to produce a weld free from ultrasonic

'indications' which would require its repair and to avoid the

existence of cracks which would also require rep air but if remaining

undetected coul d act as the nucleus for fracture- proce sses .


26/95

- 95 -

(4) Defects in the weld metal

The welding process can lead to a variety of defects in the weld

metal or adjacent HAZ of the parent materi al. Much is known about

the mechanisms of formation of these defects and how to avoid o r

minimise their occurrence by control of material composition and

fabrication procedure.

There are three forms of cracking that are potential problems in

weld deposits made in low alloy steels; solidification cracking,

reheat cracking and hydrogen induced cracking. There are no

published reports of incidences in nuclear vess el fabrication of the

first two forms , indicating that the consumables and pro cedures

normally selected in European and American nucle ar fabrication shops

have adequate resistance to these types of cracking. The

metallurgical factors controlling solidificat ion and reheat cracking

in weld deposits are reasonably well understood [23, 2 4 ] .

Hydrogen induced cracking can occasionally b e found

i n .

multi-pass

deposits in situations where the welding proce dure an d shop floor

storage and handling of consumables are not sufficiently controlled.

The cracks occur typically in arrays and are often transverse to the

line of the wel d and inclined at about 45* to the weld surface.

Sometimes referred to as chevr on cracks , the individual cracks hav e

a. zigzag appearance and c an b e transgranular or intergranular with

respect to the microstructure. They may be up to 40 mm in their


27/95

- 96 -

longest dimens ion, but are often confined to a single weld pass and,

in these situations, are less than about 5 mm in length. The

factors controlling this form of cracking are similar to those for

hydro gen cracking in the HA Z, with cracking being more likely in

situatio ns w here the weld metal hydrogen content is high, the

restr aint is high and the weld deposit microstructure is

susce ptible, this normally means that the deposit is of high

hardness

[ 2 5 - 2 8 ] .

However, cracking can occur in weld metals at

significant ly lower hardnesses than would be associated with

crackin g in the HAZ . The relationship between weld hardness,

micro stru cture and susceptibility to cracking is the subject of

current resear ch but there is sufficient knowledge at present time

to s pecify adequate procedures for fabrication in order to avoid

cracking.

For nuclea r ves sels , weld metal hydrogen induced cracking must b e

regarde d as a potential pro blem in submerged arc and manual metal

arc weld depo sits . For welds of the former type, the type of

submer ged arc flux used is an important factor to be considered,

since there is evidence that agglomerated flux types have been

assoc iated w ith a greater tendency to weld metal hydrogen cracking

than fused fluxes [2 9]. Crack-free welds can be produced with

either fused or agglomerated fluxes but the latter type needs mor e

careful sho p floor control. As stated the choice of a fused flux,

to pro vide a goo d resistance to weld metal hydrogen cracking, bring s

with it the penalty of a generally lower weld metal notch toughness,


28/95

- 97 -

due principally to the higher non-metallic inclusion content of such

deposits w he n compared with those made with agglomerated fluxes

which are mo re bas ic. It must also be noted that the level of

hydrogen in th e weld metal is partly determined by wire cleanliness

and thus wir e quality as well as flux quality is important.

4.1.2 Heat Affected Zone

(1) Defects in the weld metal

As for the wel d metal, there are three forms of cracking that are

potential pro blems in the HA Z, namely liquation cracking, r eheat

cracking and hydr ogen induced cracking.

(a) Liquatio n cracking

Liquation o r ho t cracking is a mode of intergranular cracking

occurring at elevated temperature in the initial welding thermal

cycle, i.e. befo re post-weld heat treatment. It is associated wi th

weak g rain bound ary zones o f reduced melting point material

containing enh anced concentrations of impurities particularly

sulphur, and o ccurs preferentially in localised regions of pos itive

segregation [30, 31 ]. Liquation cracking has been observed in the

HA Z of 22 NiMoC r 37 (similar to SA-508 Class 2) and in 20 MnMoNi 55

(similar to SA-5 08 Class 3 and SA-533B Class 1 ) but none has bee n


29/95

- 98 -

reported for weld metal itself

[ 3 1 - 3 4 ] .

In examinatio n of 120

weldments from test plates, production weld prolongati ons and actual

components, about 3 0 * contained liquation crac ks w ith dimensions

typically of a few grain diameters (i.e.< 1 m m ) .

The control of liquation cracking is primar ily a question of control

of bulk and local purity, because other paramet ers such as welding

technique and heat input appear to be of s econ d or der significance

[ 3 0 ] .

Restrictions on copper, tin, phos phor us, sulphur and arsenic

bulk impurity content to the levels given in Ta ble 6 have been

proposed [33] together with a 'threshold crackin g criter ion

1

such

that cracking is likely to occur if two or mo re elements exceed

these value s. Whilst this criterion success fully characterises the

cracking susceptibility of the weldments ex amine d, it must b e

conceded that local regions of segregation may occasionally lead to

small isolated liquation cracks in material satis fying these bulk

compositional requirements. Liquation crac king susceptibility as a

phenomenon generic to weldable steels, is kno wn to decrease with

increasing manganese-sulphur ratios [ 35 ]. Alt hou gh all modern PWR

pressure vessel steels, whether of the SA-533B/S A-508 Class 3 or SA-

508 Class 2 type, have high manganese-sulphur rati os, the former

typically contains

1.6-1.8

times the manganese of the latter and

therefore for a fixed sulphur impurity conte nt it may be anticipated

to be less susceptible to this mode of cra ckin g.


30/95

-

99

-

(b) Reheat cracking

Of the embrittlement and cracking phen omen a resu lting from the

fabrication of nuclear pressure vessel s, reheat cracking is

considered the most difficult to solve. Furthermor e, in spite of

extensive research of this phenomenon in PWR press ure vessel [9, 30 ,

33-49] and other steels, a complete understanding of all controlling

parameters has not yet been achieved, Howe ver, there is general

agreement in qualitative terms of the mechanis m involved and

susceptibility of individual classes of s teels ; for example SA-533B

Class 1 and SA-508 Class 3 although not immune are less susceptible

to this mode o f cracking than SA-508 Class 2.

Reheat cracks vary in size from a single grain (—20 ^ m ) to a

significant fraction of the weldment (10-im) [3 4, 3 5 ] . Cracking

occurs preferentially in regions of alloying and impurity element

segregation. Microcracks are restricted to the coarse-grained

unrefined regions of the HA Z whereas macrocra cks link coarse-grained

regions interconnected by partially refined regions of HA Z

microstr ucture. The positions and types of cracking found in the

extensive investigations by Kussmaul an d co-workers [30, 32-34] are

summarised in Fig. 30.

Reheat cracking is a high temperature gr ain boundary fracture

phenomenon occurring at temperatures bel ow about 700°c • The

occurrence is more likely in materials with coarse austenite gr ain


31/95

- 100 -

size and rel atively high levels of the impurities together with a

high hardnes s before reheating or stress relieving and a resistance

to softening at elevated temperatures. Thus the extent of cracking

in heat affec ted zone is reduced by the presence of aluminium

nitride which induces a fine grain size and prevents grain growth by

minimising t he heat input into the weld to reduce the size of the

H A Z , and by restricting levels of those elements which form fine

dispersions of stable carbides and hence are responsible for

resistance to softening during the tempering heat treatment. For

the latter pur pos e it is important to achieve very low levels of

vanadium, zirconium and niobium.

Reheat crac king results wh en the relaxation strain exceeds the local

creep ducti lity of the material . It occurs during postweld heat

treatment (PWHT) when the welded structure is heated slowly from

room temper ature or the post-welding temperature (up to 300 °c) to a

temperature between 550 and 650 °c , held at this temperature for

several ho urs and slowly cooled to minimise further residual

stresses. Cracking can occur during heat-up or holding when the

instantaneous conditions o f residual stress, hardness, accumulated

strain, micro str uctur e and interfacial segregation of impurity

elements a re consistent wi th the requirements of a particular

failure mechanism.

The creep proc esse s occurr ing during relief of residual stresses are

sensitive t o small changes in alloy and impurity element composition

and to microstructure [31, 33, 34, 41, 44, 46,


32/95

- 101 -

4 7 ,

5 0 ] . Whilst there is general agreement that SA-533B Class 1 and

SA-5 08 Cla ss 3 are less susceptible to reheat cracking than SA-508

Clas s 2 an d that impurity elements are deleterious, there is only a

broad quantitative consensus on the relative effects of different

impurities. From studies on simulated HAZs in experimental steels

contai ning single impurities, Brear and King [47] recommend that

individual elements should not exceed the limits given in Table 7

and the combined impurity element (nt.%) should be

P + 0.81 As + 1.18 S n + 1.49 Sb + 0.12 Cu + 0.195 S < 0.03

in order to avoid reheat cracking. Specimens such as those used in

this study, with simulated coarse-grained microstructure across the

whole cross-sectio n, in general , yield pessimistic results in

compariso n with those obtained from actual weldments. The results

also refer to specific test conditions and it is not clear how the y

relate quantitatively to the occurrence of reheat cracking in

service . Kussmaul et al. report an increasing tendency for cracking

with increasing contents of phosphorus, sulphur, copper, arsenic,

aluminium, nitro gen, molybdenum and cobalt [33 ]. They also propose

a 'threshold value

1

criterio n such that reheat cracking occurs if

two or mo re elements exceed the limits given in Table 6. More

recently [3 8] , examination of reheat cracking in commercial casts of

SA-533B C las s 1 and SA-508 Class 2 steels has indicated that

increasi ng the chromium content is deleter ious. Similarly the

importance of chromium and other strong carbide-forming elements

such as m olybdenum, vanadium, titanium and niobium in prompting


33/95

- 102 -

rehea t cracking in other low alloy steels has bee n recog nised [50,

5 1 ] .

(c) Hydro gen cracking

Hydro gen cracking is a potential source for d efects associated wi th

the HAZ of structural weld. This is a brittle crac king mechanism

occurring below about 200°c. The phenomenon is associat ed

particularly with high strength steels in the as-welded condition

wher e hydrog en has been introduced during welding and high levels of

stress rem ain [53]. Although no failures or large defects in PWR

plant have be en reported to be caused by hydrogen cracking, recent

wor k h as demonstrated that regions of alloy segregati on in PWR

pressu re vessels and other steels are more susceptib le to hydrog en

cracking than the matrix [ 4 9 , 5 4 - 5 6 ] . Liquation cracks , also

occurring in segregated material, are ideally suited to act as

stress concentrators for subsequent hydrogen c rackin g.

The critical concentration of hydrogen below w hich crack initiation

will not o ccur is not known, although values as low as 1.3 ppm hav e

bee n suggested [56]. A low hydrogen content in plate s and forgings

is achieved by vacuum degassing prior to casting (2-3 ppm) together

wit h one or sore heat treatments during fabrication (1 p pm ). O n

welding, the local hydrogen concentration of the weldmen t will

increase.

Wit h good welding practice, as specified in procedures

for fabrication of nuclear vessels, a concentration of about 5 pp m


34/95

- 103 -

ca n be expected. Immediately after welding the HAZ is susceptible

to hydro gen cracking and therefore should be maintained at a n

elevated temperature until sufficient hydrogen has diffused

away.Tests have shown that no cracking occurs in SA-508 Class 3

weldments provided that either a 200 °c preheat and post-heat

temperature is employed or a lower preheat temperature combined w ith

a post-heating cycle after welding is used [3 7]. Further reductions

in hydrogen concentration will occur during subsequent stress relie f

annealing.

4.2 Cladding

(1) Material and welding procedure [2]

Th e inner surfac e of the vessei is clad with a corrosion resistant

layer by continuously melt ing cladding material onto the vessel

surface to produce a fusion weld. The method is mechanised for the

singl e curvatur e surfaces in order to provide a layer of constant

thickness. Certain regions of double curvature, however, must be

clad manually. Two types of feed material are used; a type 309/308

austenitic stainless steel which is used for cladding the ferritic

stee l of the ma in pressure vessel and Inconel overlay which is use d

o n penetrations , core support pads and on the faces of the nozzl es.

Detai ls of established m etho ds of clad welding, electrode and flux

compositio ns and operating conditions are given in the review of

overlay welding by Gooch [5 7] . A recent process has been developed


35/95

- 104 -

reportedly capable of depositing high quality austenitic cladding up

to 300 nun with [5 8] .

(2) Defe cts associated wit h cladding [2]

There are several differences between structura l weld ing and

cladding which are relevant to the mechani sms of formatio n of

welding defects. In the case of cladding the depos it is mainly

austenitic rather than ferritic, and therefo re poss esses a higher

coefficient of thermal expansion and greater hig h temperature

strength, both factors influencing residual stre ss fo rmation and

stress relaxation behaviour. Also heat input is high during

cladding thus promoting residual stresses and large heat affected

zones.

Howeve r, with the exception of the ins ide of nozzles the

cladding process is a low-restraint weld co nfigur ation and th erefore

less severe with respect to long range resid ual stress fo rmation.

Hydrogen ha s a higher solubility but lower diffusivity in austenitic

than in ferritic steels. Consequently, the austenitic cladding can

retain hydrogen which may subsequently pass into the base material.

Most repor ted defects associated with the cladd ing of PWR p ressu re

vessels are related to the HAZ in the ferritic steel below the

cladding. However, defects can occur both in the cladding itself

and along the cladding/ferrite interface. Goo ch reports three

instances of cracking in the cladding duri ng fabricatio n due to

failure to achieve the desirable compos ition an d microstructur e

[ 5 7 ] .


36/95

- 105 -

(a) Underclad reheat cracking [2]

A potential problem associated with cladding of r eactor pressure

vessels is the formation of underclad cracks . The first report of

defects in the HAZ beneath austenitic cladding in nuclear plan t w as

in 1970 [59 ]. The Welding Research Council undertook a

comprehensive review of the phenomenon and Vinckier and Pense

reported this work in 1974 [ 9] . Cracks were found exclusively alon g

prior austenite grain boundaries with sizes varying from a minimum

of 0.2 mm in depth and length to a maximum of 10 mm length and 3 mm

depth. In mor e recent reviews [31, 44] the maximum depth is

reported as 4 mm. The cracks exist in a region which is somewhat

difficult to examine by conventional ultrasonic testing techniques

because of the proximity of the cladding and the surface. Cracks

have been revealed by stripping the cladding and using surface cr ack

detection method s. Metal lographic examination showed that cracks

wer e in the coarse-grained region of the HAZ which ha d been fully

austenitised by the first cladding deposit and then heated to ju st

below the austenitisation temperature, i.e. 600-700°c, by the

subsequent adjacent cladding deposit. The susceptible region is

under the highes t residual tensile stress immediately after welding .

The direction of cracking wa s usually between 45 and 90* to th e

direction of welding. Fig . 31 illustrates the positi on of the

cracks.


37/95

- 106 -

Vinckier an d Pe nse concluded that pressur e vessel steels

manufactur ed t o the different specifications have different

susceptibil ities to underclad cracking. Out of 96 reports showing

26 cases o f u nderclad cracking, 25 wer e in SA-508 Class 2, one in SA-

508 Class 3 and no cases were reported for SA-533B Class 1. High

heat input duri ng cladding resulted in underclad cracking in SA-508

Class 2 but not in the other steels. This pattern of behaviour was

confirmed b y other reviews in the period 1974-1978 [31, 44 , 60 , 6 1 ] .

The cr ackin g reported was all attributed to reheat cracking

occasi onally augmented by liquation cracking.

Figs.

32 an d 33 show reheat cracking susceptible areas and methods

to avoid reheat cracking by refining heat affected coarse grain

zones [6 2] .

(b) Under clad hydrogen cracking [54]

Hydrogen cr ackin g is one of the most important problems for the

integrity of steel structures and man y studies on this subject have

been extens ively carried out until now . The dominant factors for

the hydr ogen cracking are summarized as follows:

- existe nce of diffusible hydrogen

- existence of tensile stress or strain

- exist ence of hydrogen embrittlement susceptible

microstrueture

- low temperature condition s (below 150°c)


38/95

- 107 -

The hyd rog en crack initiates when all four factors are

simult aneou sly satisfied. In the case of welding between low alloy

ferriti c m et als , the weld joint becomes quite susceptible to the

hydr ogen crack ing and almost all the studies have been conducted o n

this su bje ct. On the other hand, weld joints of ferritic metal and

austen itic wel d metal are considered to be resistant to hydrogen

crac king b ecau se of the high hydrogen solubility, low hydrogen

diff usio n r at e in the weld metal and sufficient capacity of

relax ation o f welding induced strain. Nevertheless, the hydrogen

cracks in t he HA Z under austenitic stainless steel overlay wer e

recent ly repo rted o n the tube sheet forging of a steam generator in

a light wa ter reactor [55, 6 3 ] . The references pointed out that the

cracks ma in ly initiate in the zones of segregation.

Th e ex ist enc e of susceptible microstructure to hydrogen

embr ittl emen t is also an important factor influencing the initiation

of hydrogen cracks.

Fou r vit al conditions to initiate the hydrogen cracking have been

examined independently for the heat affected zone under the

auste nitic stainless steel overlay. From the results, it is

concluded that :

- 2 t o 4 ppm hydrogen diffuses from the austenitic weld

me ta l into the bas e metal by the «p-transformation of th e

ba se metal due to the welding

- he at affected zo ne is quite hydrogen embrittlement

susceptibl e and and the embrittlement is


39/95

- 108 -

especially noticeable at room temperature. The r emarka ble

hydrogen embrittlement occurs at hydrogen content of 1.5

ppm and higher

- segregation, which is difficult to avoid in large

forgings at the present time, is the most susce ptible to

hydro gen embrittlement

- maximum magnitude of residual stress amo unt to 500 MP a an d

the magnitude is sufficient to initiate the h ydro gen

cracking

From t he above facts, it is quantitatively proved that hy dro gen

cracking occu rs even in the weld joint between the austen itic we ld

metal an d ferritic metal.

The preventive measures against hydrogen cracking a re s ummarized as

follows. On the assumption that the materials and desi gn of

component ar e not changed for this purpose, the exist ence of

hydrogen embrittlement susceptible microstructure and restr aint

condition of weld joint cannot be avoided. On the other hand, t he

large decrease s in residual stresses under the weld over lay cannot

be expected by the conventional soaking treatm ents. Hen ce, the

countermeasur es against the hydrogen cracking become s as fo llo ws.

- elimination of hydrogen in the heat affected zone

- to avoid the low temperature conditions un der the hydr oge n

abs orb ed conditions of heat affected zone

Therefo re, it is recommended that the preheating should be

maintained a t least during 1st and 2nd layer welding and until pos t


40/95

- 109 -

wel d heat treatment or soaking treatment. The soaking treatment at

about 250°c is undoubtedly effective to avoid the hydro gen

cracking. By these countermeasures, hydr ogen cracking in the HAZ of

heavy forging with stainless steel overlay cla d can be complet ely

avoided.

4.3 Mechanical Properties of Weld Metal and Heat Affected Zo ne

(HAZ) [2]

Tensile and Charpy V-notch impact properties for the European weld

metals are comparable with the base metal data . It would appear

that adequate low temperature notch toughness and comparable tensile

ductility can be achieved in weld metals even when the weld metal

yield strength is approximately 10-20& higher than the nominal mean

value o f 470 MPa for SA-533 Grade B Class l/SA-508 Class 3 ba se

materials. Data for base metals, weld metals and HAZs which were

tested in the EPRI programmes [20] are collated in Tables 8, 4 and 9

to 1 2. Dat a from Japan which provide some indication of the

improvement in upper shelf notch toughness a nd also in the ductile -

brittle transition temperature that can be o btained using nar row gap

welding processes [64, 6 5 ].

From the results examined for both weld metal s and HAZs, which are

rather limited in some cases, it can be concluded that it is

possible to achieve mechanical properties in weld metals and he at

affected zones


41/95

- 110 -

that are at least as good as those of plates and forgings. It is

believed that the weld metal rather than the HAZ will govern

acceptance. By correct selection of welding consumables/

parameters, weld metal properties can be achieved which are well in

excess of the minimum property values specified by the ASME

C o d e s .

The upper shel f notch toughness of weld metals should also be

reasonably high , but cannot be guaranteed to match the high notch

toughness values of the very high quality steel now available from a

number of sour ces. Recent developments in the production of plates

and forgings , and to a lesser extent w e l d s , have resulted in an

improvement in properties, beyond the minimum values required by the

A S M E C o d e s , and therefore beyond the material properties upon which

p r e v i o us P W R v e s s e l p r o d u c t i o n h a s b e e n b a s e d .

5 A d v a nc e d D e s i g n d u e t o O p t im i z ed M a t e r i a l

The design of the RPV for the light water reacto r (LWR) tends to

m i n i m i ze t h e w e l d s e a m s , which reduces the period of in-service

inspection (ISI) together with easier perfo rmance of ISI. This

tendency reg uired the more integrated and larger pa rts for nuclear

steam supply system (NSSS) components.

It was said t hat weld seams can be reduced to 70 percent for boiling

water reactor pres sure vessels (BWRPVs) and 25 percent for

pressurized wate r reactor pressure vessels (PWRPVs) when compared to


42/95

- Ill -

conventional desig ns, by the use of the large forgings and plates

available at present in the world [6 6] . Typical layouts for these

designs of nu clear pressure vessels are as shown in Fig.34 for the

P W R a n d F i g . 3 5 f o r B W R.

The seamles s forged shells for the BWRPV and PWRPV, as well as the

vessel flang e integral with nozzle belt shell in PWRPV, are

significantly advantageous from the standpoint of design,

fabrication and inspection.

The nume rou s seamless for ged shells in BWRPV and PWRPV have already

b e e n r e a l i z e d a s s h o w n i n F i g s . 3 6 a n d 3 7 [ 6 7 ] , r e sp ec ti ve ly , u s i n g

the advan ced technology for the manufacture of heavy steel forgings.

One-piece fo rg ed shell flanges weighing 165 tons for KWU type

4-loop

P W R P V m a d e f r o m 4 0 0 t o n in g o t s s h o w n i n

Fig.

1 7 h a v e b e e n ~

s u c c e s s f u l l y d e v e l o p ed [ 1 3 ] . F u r t h er m o r e , o n t h e b a s is of m u c h

manufa cturi ng experience of one-piece shell flange for KWU, mo no -

block ves sel flange of WEC t ype 157 " (3988 mm ) PWRPV, combined

v e s s e l f l a n g e w i t h n o z z l e b e l t s h el l a n d o n w h i c h t he s e t -o n t y p e

m a i n c o o l a n t n o z z l e s a r e w e l d e d a s s h o wn i n

F i g .

1 8, was manufactured

s u c c e s s f u l l y u s i n g a 5 0 0 t o n i ng ot [6 8 - 7 1 ] .


43/95

- 112 -

REFERENCES

[1] ON OD ER A, S., Tetsu To Hagane, Current Steels for Nuclear Pressure Ves sels, J.

ISIJ (Iron and Steel Institute of Japan), 67 (1981), P880

[2] Second Rep ort on Assessment of the Integrity of PW R Pressure Ves sels, UK.AEA

(Mar. 1982)

[3] STEE LE, L.E., Neutron Irradiation Emb rittlement of Reactor Pressure Vessel

Steels, IAEA (1975)

[4] AN DO , Y., Private Com munication to Mr. Steele, L.E. (21 No v. and 10 Dec .

1973)

[5] HAS HIMO TO, U., KIHARA, H., AND O, Y., Welding Problems Associated with

Construction of Nuclear Power Stations in Japan, 1964 Annual Assembly of IIW

(International Institute of Welding), Prague, Czechoslovakia (30 Jun. - 2 Jul. 1964).

Proceedings

[6] RA O, S., Private Com munication to Mr. Steele, L.E. (14 No v. 1973)

[7] PA CH UR , D., Private Com munication to Mr. Steele, L.E. (11 No v. 1973)


44/95

- 113 -

[8] DE BR AY , W., CERJAK , H., Werkstoffeigenschaften des Stahles 22 NiM oC r 37

fur Reaktorkomponenten, VGB-Werkstofftagung (1971)

[9] VIN CK IER, A.G., PENSE, A.W ., A Review of Underclad Crack ing in Pressure

Vessel Components, WRC Bulletin 197 (Aug. 1974)

[10] CER JAK , H., DEB RAY , W., PAPOU SCHE K, F., Eigenschaften des Stahles 20

MnMoNi 55 fur Kernreaktor-Komponenten, VGB-Konferenz, Werkstoffe und

Schw eisstechnik in Kraftwerk 1976 (197 6), Diisseldorf

[11] KUS SMA UL, L., EWA LD, J., MAIER, G., SCHE LLHA MM ER, W., Enhancement

of the Quality of the Reactor Pressure Vessel Used in Light Water Power Plants by

Advanced Material, Fabrication and Testing Technologies (Aug. 1977), San

Francisco

[12] ON OD ERA , S., FUJIOKA , K., TSUK ADA , H., SUZ UK I, K., Ma terial

Specifications towards more Reliable Fabrication of Nuclear Pressure Vessels, The

2nd Joint Symposium of TUV Rheinland and B & W/USA (28 Sept. 1978), Koln.

Preprint

[13] DEB RAY , W., CERJAK, H., ONODERA , S., TSUK ADA , H., SUZ UKI, K., Large

Nozzle Belt Forgings for PWR 1200 MWe, 1st European Nuclear Conference

(April 1975), Paris. Preprint


45/95

- 114 -

[14] U.S. NRC Regu latory Guide 1.99, Rev. 1, Effect of Residual Element on Predicted

Radiation Damage to Reactor Vessel Materials (Apr. 1977)

[15] JSW (Japan Steel W orks), In-C omp any Document (1 June 1982)

[16] ONODERA , S., FUJIOK A, K., TSUK ADA , H., SUZUK I, K., Advantages in

Application of Integrated Flange Forgings for Reactor Vessels; The 3rd MPA

Seminar (15 Sept. 1977), Stuttgart, MPA, University of Stuttgart, Germany, 1978

[17] COU PETTE, W., Der Einfluss der Seigerung und Verschm iedung auf die

Festigkeitseigenschaften grosser Schmiedestucke aus Stahl, Stahl und Eisen, 61

(1941),

P1013

[18] WEL FLE, K., BITT ER SM AN N, H., Flockenfreigliihen mit

Zwischenstufenumwandlung, Neue Hiitte, 11 (1966), P730

[19] BRU CKN ER, E., et al., The Properties of Various Weld Metals for Reactor

Components, DVS Berichte No. 52 Welding in Nuclear Engineering 1978 67-73.

V.E. Riecansky, Technical Translation No. VR/1318/78, Cambridge

[20] Nuclear Pressure Vessel Steel Data Base . EPR I NP9 33 Proj. 886-1 (December

1978)

[21] HAW THO RNE , J.R., W elding J., 51 369S (1972)


46/95

-

115

-

[22] DAVIS, M.L.E., BAILEY, N., How Submerged Arc Flux Composition

Influences Element Transfer, Proceedings of Conference on Weld

Pool Ch emis try an d Metallurgy, The Weldi ng Institute (April

1980)

[23] BAILEY, N., JONES, S.B., Solidification Cracking of Ferritic

Steels dur ing Submerged Arc Welding, The Welding Institute

(1977)

[24] BATTE, A. D, MURPHY, M.C., Reheat Cracking in 2 Cr/Mo Weld

Metal : Influence of Residual Elements and Microstructure,

Metals Technology, Vol.6, No.2 (February

1 9 7 9 ) ,

P62

[25] HART, P.H.M., Weld Meta l Hydrogen Cracking, Welding Institute

Research Bulletin (November

1 9 7 8 ) .

[26] KEV ILLE , B.R., An Investigation to Determine the Mechanism

Involved in the Formatio n and Propagation of Chevron Cracks in

Submer ged A rc Weld ment s, Welding Research International 6 (6)

(1976)

[27] MOTA, J.M.F.,

APPS,

R.L., JUBB, J.E.M., Chevron Cracking in

Manual Met al Arc Welding, Proceedings, Trends in Consumables

and Ste el s for Weldin g, The Welding Institute, (November

1 9 7 8 ) , London


47/95


48/95

- 117 -

[33] KUSSMAUL, K., EWALD, J., et al., Enhancement of the Quality of the Reactor

Pressure Vessel Used

in

Light Water Power Plants

by

Advanced Material

Fabrication and Testing Technologies; 4th Int. Conf. on Structural Mechanics in

Reactor Technology; Paper Gl/3 (August 1977), San Francisco

[34] KUSSMAUL, K., EWALD, J., Assessment of Toughness and Cracking in the Heat

Affected Zone

of

Light Water Reactor Components;

3rd Int. Conf. on

Pressure

Vessel Tech., Part II, P.627-646, ASME (1977), Tokyo

[35] Solidification Cracking of Ferritic Steels during Submerged Arc Welding; The

Welding Institute (1977)

[36] BREAR, J.M. and KING, B.L., Phil. T rans. Roy. Soc, London A295 (1980)

[37] COMON, J., A508 Class 3 Forgings for Pressure Vessels; 3rd Int. Conf. on

Pressure Vessel Technology, Part

II,

P.957-970, (April 1977), Tokyo, Japan.

Atomic Energy Society of Japan, 1977

[38] McMAHON, C.J., DOBBS, R.J., GENTNER, D.H., Stress Relief Cracking in

MnMoNi and MnMoNiCr Pressure Vessel Steels, Mat. Sci. and Eng. 37 (1979),

P. 176-186


49/95


50/95


51/95

- 120 -

[49] TSUKAD A, H., TAN AKA , Y., OHN ISHI, K., Temperature Dependence of

Hydrogen Embrittlement of Weld Thermal Cycled Material; Japanese Welding Jnl.

[50] VIEILL AR D- BA RO N, B ., Deve lopm ent of the Production of Special Steels for

Nuclear Industries, Materiaux et Techniques (January 1977), P321-337

[51] ITO, Y., NA KA NIS HI, M ., IIW (International Institute of W elding), Doc. X- 668-

72

[52] NAK AMU RA, N., NA IKI, T., OKA BAY ASHI, H., Fracture in the Process of

Stress Relexation under Constant Strain; Proc. First Int. Conf. on Fracture (1965),

Sendai, Japan, P863-878

[53] JPVRG Report N o. 2, Tem per Em brittlement and Hydrogen Embrittlement in

Pressure Vessel Steels; Iron and Steel Inst. of japan (May 1979)

[54] OHN ISHI, K., TSUK ADA , H., SUZUK I, K., MU RAI, H., KAG A, H.,

KUSUHASHI, M., OGAWA, T., TANAKA, Y., Study on Hydrogen Induced

Cracking of Heavy Forgings Overlaid by Stainless Steel; The Special Topical

Meeting, Metal Performance in Nuclear Steam Generators; ANS (6-9 Oct. 1980),

St. Petersburg Beach, Florida


52/95

- 121 -

[55] CER JAK, H., SC HM IDT , J., and LOH NB ERG , R., Correlation between

Segregation and Cold-Cracking; 4th MPA Seminar on Safety of Pressurized LWR

Containments (October 1978), Risley, Tran.4253

[56] WID AR T, J., Private Com mun ication

[57] GO OC H, T.G., Re view of Overlay Welding Procedures for Light Water Reactor

Pressure Vessels; Welding Institute Document 3455/1/75 (1976)

[58] Kawasaki Steel Corp oration, Mag lay; The Ne w Surfacing Process with Electroslag

Welding using Wide Strip Electrode

[59] W YLIE , R.D., Rep ort of PVR C Task Group on Underclad Cracking, HSST (Heavy

Section Steel Technology); 6th Annual Information Meeting (April 1972), ORNL

Conf. 720468

[60] DO LB Y, R.E., and SAU ND ER S, G.G., Underclad Cracking in Nuclear Vessel

Steels, Part I, Occu rrence and Mech anism of Cracking. Met. Constr. (December

1977),

P562-566

[61] DO LB Y, R.E. and SAU ND ER S, G.G., Underclad Cracking in Nuclear Vessel

Steels, Part II, Detection and Control of Underclad Cracking. Metal Constr.,

(January 1978), P20-24

[62] VIG NE S, A., Private Com mun ication to Dr. Onodera, S. (Jan. 1983)

[63] Cracks in French Pressu re Vessels Pose N o Danger; Nuclear Engineering

International (Jan. 1980), P27


53/95

- 122 -

[64] SAW ADA , S., OHT A, M., NISHIOKA, K., HORI, M., KAW AHA RA, M.,

YAMASAKI, H., Application of Narrow-Gap GMA Welding Process to Nuclear

Reactor Pressure Vessels; presented 4th Int. Conf. on Pressure Vessel Technology,

PI 13-1 19, Vol. II, I. Mech. E. (May 1980), London

[65] MO RIG AK I, O., et al., Submerged Arc Narrow-G ap Welding Process with One

Run per Layer Technique for Heavy Sections; Nippon Steel Welding Products and

Engineering Co. Ltd. (February 1979(, IIW (International Institute of Welding),

Doc. XII-A-168-79

[66]

ISHIKA W A, K., AN DO , E., Experience Leads to Major Advances in Reactor

Pressure Vessel Design; Nuclear Engineering International (July 1977)

[67] JSW (Japan Steel Wo rks, Ltd.); In-Company Document

[68] ONO DERA , S., MO RITANI, H., TSUCHIYA, K., TSUKADA , H.

5

WIDART, J.,

SCAILTEUR, A., Mono-Block Vessel Flange Forgings for PWRPV 1000 MWe;

the 8th International Forgemasters Meeting (October 1977), Kyoto

[69] CA M BIEN , R.B ., New Design of PWR Reactor Vessel Using Large Forging ;

ASME Conference (September 1976), Mexico City

[70] RE YN EN , J., W IND T, P.DE., WIDA RT, J., et al., A Novel Design for LW R

Pressure Vessel Nozzles and Corresponding Stress Analysis; 3rd International

Conference on Pressure Vessel Technology (April 1977), Tokyo, Japan. Atom ic

Energy Society of Japan, 1977

[71] W IDA RT, J., Design of Reactor Pressure Vessel Considering Easier I.S.I.; IAEA

Technical Committee (April 1977), Kobe


54/95

- 123 -

c

c

o

a.

£

o

o

a:

a)

CD

•*—

V )

o

_

Z3

o

1 _

CO

t

e

s

i_

o

o

rv

O

a

r—

-

/

/

CM

v_

3

O

CD

e

m

p

1

-F

Is

1 .

Is

1

s

t

e

m

f

E

o

o

cc

>

>•

O

t r

O

O .

£

o

o

o

o

E

a >

JC

O

1

or £ ;

—* -5

°°.

^ . E

l - ^ E

?

tij c

f- •>= E

•

n

\< v

<

>

o

r ^

cn

OL

•

..

x ri

{_ >

— o

o

( _ • » -

PI — -

£ O C

O Q. O

/

/

/

/

/

/

, i

I

, 1

00

A l * ~

in

Ai . .$

i •

J

53- CD

d d

to

1

^r o

o o

on g

v i o

- 8

to o

v i d

1

1

1

5

* $

o <

1

Oo

no

0 0

< » —

Pi

I D

i

to

o

V I O

to °

*

«i

d o

to

CM

V I O

^ - o

^ o

o —

to

C M

o

v i d

- » I O

^ • C M

^ O

v i d

1

2

1

5

t o

C M

- v id

to

C

l

CO

o

to

CO

c>

a

o>

o

. 1

1

i

C O

AJ g

A i ^

AI

th

0 -

AI

o

VI C5

ft

o

CM

v i o

^ o

r> C D

^ d

3

VI O

^ S4

o

vi - O

^ to

0

1

0

3

to

—

~ ro

1

o i

° - JE

> -.

__

o < r

E Cb

a

e

o

c

• a

or

c

c

LiJ

O *3

no

V - CO

1

Q) O

V l

£ CO

c a

^

— o

£ o

c rO

In

_ O

o

«~?

v t

O

E O

a

or

or

J C

cu

]O

> -

en

— ' C M

t o

o

ro

ro

to

<

CO

CO

<

fo

CM

q

o

V I

a .

E

^>

Q .

C L

3

en

CO

O

LO

<

C O

m

CO

<

CM

o

•

o

I

a .


55/95

- 124 -

TABLE 2

M in im um M e c h a n i c a l P r o p e r t i e s S p e c i f i e d i n t h e ASME C o de s ( 2 )

A 53 3B 1 P l a t e s A 50 8 3 F o r g i n g s

T e n s i l e 2 0 *C 3 5 0 ' C 2 0

8

C 350°C

Yield stress (MPa)

U l t i m a t e t e n s i l e

stress (MPa)

El ( in 50 mm) %

R of A %

Charpy Impact

Energy (J)

Lateral expansion (mm)

Minimum ave. value (+)

of three specimens

Minimum value of one

specimen

345 285(*)

552 527

18

-

68 J a t

0 .8 9 mm

X

X

34 5

550

18

38

RT

ND T

+ 33"

at RT

ND T

+

41 J a t

34 J a t

285

-

-

C

33°C

4.4°C

4.4*C

("*) non-mandatory

not more than

this value

to be specified by purchaser

(*) no t more than one specimen from a se t may fa ll below

this value


56/95


57/95

- 126 -

As-Deposited

Heat C

Chemical

Mn

Analyses

P

TABLE

of EPR1

iwt.

S

4

. ' .SO Heat

1

" ( 2 0 )

S i Ni

Weld Metal

•

Cr

s in

Mo

A533B

Y

Cl 1

Cu

MMA

Welds

P

Q

R

S

T

U

S/A Welds

V

W

X

Y

Base

~B~~

L

N

0.100

0.100

0.100

0.09

0.04

0.050

0.150

0.14

0.150

0.13

0.25

0.21

0.24

1.000

1.110

1.100

1.03

1.02

0.150

1.38

1.19

1.280

1.25

1.41

1.34

1.30

0.005

0.007

0.006

0.005

0.017

0.016

0.008

0.01

0.011

0.011

0.008

.0.012

0.009

0.010

0.010

0.010

0.01

0.022

0.024

0.009

0.009

0.010

0.010

0.014

0.019

0.013

0.390

0.400

0.400

0.39

0.49

0.520

0.16

0.19

0.200

0.18

0.260

0.230

0.240

0.880

1.060

1.000

0.95

0.95

0.940

0.13

0.10

0.190

0.10

0.46

0.44

0.46

0.010

0.010

0.010

0.01

0.01

0.010

0.04

0.09

0.080

0.09

0.11

0.07

0.11

0.290

0.340

0.330

0.32

0.53

0.540

0.60

0.54

0.540

0.53

0.49

0.53

0.53

0

0

0

0

0

0

0

0

0

0

0

0

0

.004

.006

.005

.006

.014

.014

.007

.005

.005

.005

.003

.004

.002

0.020

0.020

0.020

0.020

0.02

0.030

0.04

0.12

0.110

0.20

0.12

0.10

0.08


58/95

TABLE 5

(

C h e m ic a l c o m p o s i t io n , o f f i l l e r me t a l

as

s p e c i f i e d ,

as

r e c e i v e d

and as

d e p o s i t e d ,

t o g e t h e r w i t h t h a t

o f t h e

base meta l (SA533B)

( 21 )

Ch e m i c a l c o m p o s i t i o n , wt .%

C

Mn P S Si Ni Cr Mo Cu V A l As Sn Sb

F i l l e r m e t a l :

s p e c 1 f 1 c a t i o n

a

. 15 1 .80 LAP LAP 0.1 0 .55 .10 .45 LAP .02 .05 LAP LAP LAP

7?0~ 2 TT0" OTOTO OTOTO Tma x)

T7 F (max) j5o~ 0T W

(inaTx)

( i a x )


59/95


60/95


61/95

TABLE. .9

Tensile P roperties

o f

EPRI Base Metals and Melds

( 2 0 )

M a t e r ia l

Tensile M-ft weld i n M f l weld in S /A weld i n M A IIAZ i n S/AI1AZ in

Prope rly A533B1 base ( T ) AS3301 ( L ) A53301 ( T ) A533B1 ( L ) A53301 ( L ) A533U1 (L )

Y i e l d s t r e s s

(f-Pa)

UTS

E l o n g a t i o n

R o f A

T'C

-24

200

-2 0

288

- 2 0

288

- 2 0

288

n

13

tl

II

•1

It

tl

' II

II

X

435

386

592

578

27.7

23.3

64.1

56 .4

lsd

2 9 . 9

25 .0

27.3

2 4 . 4

2. 0

2. 1

3.2

5.5

T'C

43

208

43

288

43

288

43

288

n 3f

4 478

" 437

" 577

" 570

11

3 0 . 3

" 27.4

" 74.6

" 6 9 . 0

ls d

4. 5

27.1

4.2

9.7

1.7

111

1.1

3.1

T'C

23

200

23

288

23

288

23

288

n

x

2 400,

556

11

4 3 4 ,

474

11

607,

632

" 500,

599

11

15 .5 ,

26 .2

" 19.0 ,

22.5

" 3 6 . 7 ,

65 3

11

6 6 . 7 ,

67.3

T'C

40

200

40

288

40

288

40

288

n

4

II

II

"

II

II

II

H

J

537

470

620

596

25.6

2 3 . 9

72 .1

65 .1

lsd

14.1

19.3

8.8

23.3

0. 5

1. 5

0.7

1.8

T'C

23/43

288

23/43

288

23/43

288

23/43

288

n

3

H

"

H

II

It

II

H

X

444

412

598

591

25.3

21.6

69.7

59.1

1st

4 7 . 9

3 9 . 8

4 4 . 8

36 .0

1.2

1.6

3.4

10.2

T'C

60

208

60

280

GO

288

60

200

n

1

it

"

M

II

It

II

II

x"

403

395

555

590

22

22

60.5

61.8

A508-2 base

( L )

m veld

in S /A

waid

i n

A508-2

( L )

A508-2

( L )

Cfy (M>a)

o \ (M>a)

El {%)

R o f A

(%)

•

. T'C

24

288

24

288

24

208

24

2 2 0

n

x

5 440

" 402

11

595

". 585

" 27.4

11

24.3

" 70.0

" 66.6

ls d

3 3 . 6

26 .7

2 3 . 5

21.2

1.0

0. 5

1.1

1.65

n S

5 526

" 443

" 611

" 567

" 27.0

11

22.9

" 71.6

11

64.4

1st

50.5

20.6

52.9

31 .1

1.0

1.6

1.4

4.5

n S

5 507

" 434

" 619

" 5(30

" 25 .3

" 21 .0

11

67.4

" 62.4

ls d

47

30.1

36 .2

32.5

1.7

1.7

4 .3

5. 5

n

» n o . o f

heats

x - mean property value

(L )

"

longitudinal orientation

(T ) « transverse orien tation


62/95


63/95


64/95

- 1 3 3 -

Charpy Impact and

Electrode

/Flux

E8O15.C3

u

II

E8O18.C3

n

MnMoNi/

Linde 8 0

HnMoNi/

Linde 0 0 9 1

Heat

Input

W/m)

Manual

f

TABLE

1 1

RT

MD T

D a t a

f o r

W e l d s

i n

A 5 0 8 - 2

(Effect

of

Heat Input)

Heat,

Weld

RT

NDT

CO

total Arc

0 . 8 G

1.2 I

2 . 2 K

0 . 7 H

2 . 3 J

Submerged Ar c

3 . 2

3 . 6

4 . 0

2 . 8

4 . 0

P

0

L

N

M

- 7 3

- 6 2

- 6 2

- 4 0

- 5 1

- 1 8

- 1 8

- 9

- 5 1

- 6 2

Metal

To C

CO

- 3 8

- 3 7

-

nuclear pressurized water reactor (pwr) - mechannical design materials

Documents