1

CONTENTS
1 . Ge ne rali t i e s
2 . Calc ulat ion of s ound powe r
3 . Pre dic t ion of ae rodynamic nois e
4 . Low noi s e c ont rol valve s
5 . Influe nc e of piping
6 . Ve nt s
7 . Ac ous t ic ins ulat ion
- Bi bli ography
2
CONVERSATION
POWER
HAMMER
TRAIN AT100 m DISTANCE
TRESHOLD OF PAIN
SO UND PRESSURE LEVEL d B( A)
SLEEPING ROOM
INSULATED
LIVING ROOM
CLOSE TO
AN AIRCRAFT
AT TAKE-OFF
MAXIMUM
POSSIBLE
NOISE
DISK SAW
GRINDING MACHINE
50 kW ELECTRICAL MOTOR SPRAY PAINTING
LATHE
N.B.: Sound pressure levels over 130 dB(A) may cause immediate damage to hearing
even for short exposures.
3
1
GENERALITIES
Th e a er odyn a mic n ois e is t h e mos t impor t a n t compon en t of t h e a cou s t ic
pr oblem of a con t rol va lve, s in ce it is gen er a t ed by t h e pr es s u r e wa ves
pr odu ced by t h e flu id t u r bu len ce or by ot h er flu odyn a mic ph en omen a
con n ect ed wit h s u per s on ic wa ves (impa ct cells ).
Ca vit a t ion a n d mech a n ica l vibr a t ion s a r e in compa r is on ju s t pot en t ia l
n ois e s ou r ces , beca u s e it is pos s ible t o a void t h em (a t lea s t t h eor et ica lly),
wh ile it is n ot pos s ible t o con t r ol a flu id flow r a t e wit h ou t gen er a t in g
t u r bu len ce.
For t h is r ea s on t h e n ois e is a lmos t ever n egligible in ca s e of n on ca vit a t in g
liqu ids , wh er e t h e velocit y is low, wh ile it is s en s ible for ga s a t s u bs on ic
con dit ion s a n d ver y lou d u n der cr it ica l flow con dit ion , wh er e velocit y a n d
t u r bu len ce become ver y h igh .
Th e a er odyn a mic n ois e of con ven t ion a l va lves h a s n ot a ch a r a ct er is t ic
acou s t ic s pect r u m wh ich can be eas ily iden t ified, s in ce it h as h igh volu mes
in a wide r a n ge of fr equ en cies bet ween 1000 a n d 8000 Hz, wit h pr eva ilin g
pea ks bet ween 2000 a n d 6000 Hz. High er fr equ en cies a r e gen er a t ed by
va lves pr ovided wit h low n ois e t r ims , wh er e r ea lized wit h ma n y s ma ll
flows a r r a n ged in pa r a llel.
Th e a cou s t ic power gen er a t ed by a flu id in t u r bu len t flow con dit ion is a
fu n ct ion of t h e mech a n ica l power Wm of t h e s t r ea m a n d is a s ma ll fr a ct ion
of it , t h e s o ca lled a cou s t ic efficien cy, gen er a lly defin ed a s :
wh er e:
Wm 1/ 2 q u
m
2

( W in wa t t , q
m
in kg/ s , u in m/ s )
In ca s e of fr eely expa n ded jet s t h e pr oblem is r a t h er s imple, beca u s e,
bes ide t h e fa ct t h a t t h er e a r e n eit h er down s t r ea m pipin g n or ot h er s h a pe
con s t r a in t s , a ll t h e mech a n ica l en er gy Wm ch a n ges t o t u r bu len ce.
For va lves , on t h e con t r a r y, s u it a ble pa r a met er s mu s t be in volved, t o
t a ke in t o a ccou n t t h e a cou s t ic a t t en u a t ion of t h e pipin g, t h e body s h a pe
a n d ma in ly t h e in complet e t r a n s for ma t ion of Wm in t o t u r bu len t flow du e
t o t h e pr es s u r e r ecover y a ft er t h e t h r ot t lin g s ect ion .
Th e mos t impor t a n t of s u ch pa r a met er s is t h e r ecover y coefficien t F
L
,
wh ich , a t s u bs on ic flow con dit ion s , r epr es en t s t h e en er gy fr a ct ion wa s t ed
in s ide t h e va lve.
Th e dia gr a m of Fig. 1 s h ows t h e en er get ic pr oces s t a kin g pla ce in s ide t h e
va lve a n d emph a s izes t h e r ole of F
L
coefficien t .
Th e en t h r opy in cr ea s e is ca u s ed by t u r bu len ce a n d fr ict ion s gen er a t ed
ma in ly down s t r ea m t h e ven a con t r a ct a .

Wa
Wm
Wm 1/ 2 q u
m
2

4
Enthropy
E
n
t
h
a
l
p
y
subsonic
condition
critical flow
condition
hypercritical
flow condition
fract.
fract.
p
vc
p
p p
F
1
1 2
L
2


p
vcc
p
2
1
/ 1
1,3
0,546 p
1 1

_
,

Fig. 1 Thermody namic balance ins ide the valve

Fig. 2 Pres s ure run ins ide a s ingle s tage control valve for
F
L
= 0,9 s ervicing w ater s team ( = cp/ cv = 1,3, p
1
=p
1
'
)
Th e en t h a lpy decr ea s e bet ween in let a n d ou t let t a kes pla ce on ly wh er e
t h e kin et ic en ergy in cr ea s es .
Fig. 2 s h ows t h e flu odyn a mic pr oces s es wh ich t a ke pla ce in s ide t h e va lve
a s a fu n ct ion of pr es s u r e a n d mor e exa ct ly a s a fu n ct ion of p
2
ch a n ges
wh ile p
1
is con s t a n t .
p p F p 1
2
1
/ 1
= 1 ,3 F 0 ,9
0 ,6 3 p
2c 1 L
2
1
L

+

_
,

1
]
1
1

;
5
Th e flu odyn a mic pr oces s es ca n be s u mma r ized a s follows :
1 . Subs onic flow c ondit ion (p
2
p
2c
, wh er e p
2c
is t h e down s t r ea m
pr es s u r e, cor r es pon din g t o t h e t r es h old of cr it icit y).
Un der t h is con dit ion pa r t of t h e mech a n ica l en ergy exis t in g in ven a
con t r a ct a is r ecover ed a s pr es s u r e en er gy down s t r ea m t h e ven a
cont ract a. The r emaining energy is was t ed by t u rbu lence, t hu s changing
in t o h ea t a n d n ois e.
2 . Crit ic al flow c ondit ion (p
2c
> p
2
p
vcc
, wh er e p
vcc
is t h e pr es s u r e
in ven a con t r act a u n der cr it ical flow con dit ion ). Un der s u ch a con dit ion
t h e flu id s peed in ven a con t r a ct a r ea ch es t h e s ou n d s peed a n d
s u per s on ic impa ct wa ves a r is e down s t r ea m. Th e mor e p
2
decr ea s es
t h e lower is t h e fr a ct ion of en er gy is oen t h r opica lly r ecover ed a n d
con ver t ed t o pr es s u r e; t h is fr a ct ion lower s down t o zer o wh er e p
2
r ea ch es t h e p
vcc
va lu e.
Un der t h is con dit ion a lou d n ois e is given ou t , du e t o t h e fa ct t h a t t h e
s ou n d velocit y is r each ed an d ot h er complex aer odyn amic dis t u r ban ces
ar e gen er a t ed.
3 . The hype rc rit ic al flow c ondit ion t a kes pla ce wh er e p
2
< p
vcc
.
Th e en erget ic mean in g of F
L
is n ot valid an y lon ger s in ce n o is oen t h r opic
pr es s u r e r ecover y t a kes pla ce.
All of t h e flu id kin et ic en er gy in ven a con t ract a is was t ed in in t er fer en ces
a mon g s u per s on ic impa ct wa ves .
6
2
CALCULATION OF ACOUSTIC POWER
Equ at ion s for calcu lat ion of a n d Wa for differ en t flow con dit ion s a r e
s u mma r ized in t h e t a ble of Fig. 3.
For mor e det a iled a n a lys is of t h is a rgu men t s ee t h e docu men t s lis t ed in
bibliogr a ph y u n der [1] [2] [3] [4] [5].
Acou s t ic efficien cy is plot t ed in Fig. 4 ver s u s p
1
/ p
2
for differ en t F
L
va lu es .
It is in t er es t in g t o r ema r k t h e pa r t icu la r depen den ce of a cou s t ic power
on r ecover y fa ct or F
L
.
Nois e t es t on 1-9111 Limiph on con t r ol va lve DN 3" x 4" ca r r ied ou t on s t ea m.
Ups t r ea m pr es s u r e = 92 ba r a bs , u ps t r ea m t emper a t u re = 485C.
Tes t s h a ve been per for med in a ccor da n ce wit h IEC 534-8-1, mea s u r in g
n ois e level in a n a n ech oic ch a mber, a t SIET SpA - Pia cen za - ITALY
7
F
i
g
.

3
A
c
o
u
s
t
i
c

e
f
f
i
c
i
e
n
c
y

a
n
d

p
o
w
e
r

f
o
r

d
i
f
f
e
r
e
n
t

f
l
o
w

c
o
n
d
i
t
i
o
n
s
F
l
o
w

c
o
n
d
i
t
i
o
n
S
u
b
s
o
n
i
c
C
r
i
t
i
c
a
l
H
y
p
e
r
c
r
i
t
i
c
a
l
c
a
u
s
e

o
f

a
e
r
o
d
y
n
a
m
i
c
n
o
i
s
e
t
u
r
b
o
l
e
n
c
e

d
o
w
n
s
t
r
e
a
m

o
f

v
.
c
.
t
u
r
b
o
l
e
n
c
e

+

s
u
p
e
r
s
o
n
i
c
i
m
p
a
c
t

w
a
v
e
s
s
u
p
e
r
s
o
n
i
c

i
m
p
a
c
t

w
a
v
e
s
p
2
p
p
c
2
2

p
v
c
c
p
p
c
2
2

<
p
2
2
p
c
p
v
c
c
p
p
v
c
c
1
2
2

<
p
p
2
2
p
c
p
v
c
c
2
1
2
<

p
v
c
p
p
p
F
1
1
2
L
2
-
-
/
/
/
p
v
c
c
/
p
2
1
1
1

_ ,
p
c
2
(
)
p
F
p
p
v
c
c
1
L
1

2
M
v
c
2
p
M
(
1
)
R
T
p
p
v
c
1
1
1
1

_ ,

1 ] 1 1 1 1

1
1
/
/
/
M

j
/
2
1
p
p
c
p
p
v
c
c
1
1
1
2
2

_ ,

1 ] 1 1 1 1
/
a
c
o
u
s
t
i
c

e
f
f
i
c
i
e
n
c
y

1
0
M
v
c
4
3
,
6

1
0
M
j

4
L
2
6
,
6

F

1
0
M
j
2
1
,
4
4
2
L
2
6
,
6

F

3
,
4
1
0
1
,
4
4
L
2
6
,
6

F

W
a

F
W
m
L
2

W
m
p
p
p
p
v
c
c
1
2
1

W
m
8
Fig. 4
Acous tic efficiency -
as a function of p
1
/ p
2
and of F
L
for = 1.3 -
R
e
n
d
i
m
e
n
t
o

a
c
u
s
t
i
c
o
A
c
o
u
s
t
i
c

e
f
f
i
c
i
e
n
c
y
Regime ipercritico - Hypercritical flow condition
Regime critico - Critical flow condition
P
1
/ P
2
9
3
PREDICTION OF AERODYNAMIC NOISE
3.1 EQUATION FOR CALCULATION
Th e a cou s t ic power Wa gen er a t ed by t h e flu id in s ide t h e va lve is
obt a in ed by mea n s of equ a t ion s s h own in Fig. 3.
For t h e ca lcu la t ion of s ou n d pr es s u r e level Lp r efer t o t h e followin g
equ a t ion :
W
p
c
2

wh er e S is t h e flow s ect ion a l a r ea of t h e s ou n d wa ve, p is t h e a cou s t ic

pr es s u r e a n d c t h e media impeda n ce.
Du e t o t h e pa r t icu la r ch a r a ct er is t ic of t h e a s s embly va lve+pipin g , t h e
a pplica t ion of t h is equ a t ion is r a t h er complex, s in ce t h e followin g fa ct or s
ar e in volved in t h e ca lcu la t ion :
1. Th e in t egr a t ion s u r fa ce of s ou n d power
2. Th e fr a ct ion of a cou s t ic power t r a n s mit t ed t o a dja cen t pipin g [5]
3. Th e fr equ en cy dis t r ibu t ion of t h e gen er a t ed n ois e [5]
4. Th e effect of flu id velocit y in s ide t h e pipin g
5. Th e a cou s t ic a t t en u a t ion of t h e pipin g
Her e it is t h e fin a l equ a t ion for t h e ca lcu la t ion of t h e s ou n d pr es s u r e
level:
Lp(A) 160 10 log Lg sp 10 log T
Wa c r
w
Di
Di 2000
Di
Lfp
2 2
2
+ + + +

+

(1)
wh er e:
Lp(A) = A-weigh t ed s ou n d pr es s u r e level, mea s u r ed a t 1 m dis t a n ce from
va lve ou t let a n d 1 m dis t a n ce fr om t h e ba r e pipe wa ll
r
w
= fr a ct ion of a cou s t ic power t r a n s mit t ed down s t r ea m - for values
s ee table of Fig. 5
Lg = cor r ect ion for down s t r ea m velocit y =
16 log
1
1 M
2

s p = cor r ect ion fa ct or of s pect r u m - s ee table of Fig. 6

Ty p e s o f PARCOL valve s rw
1-69 51; 1 -69 21 ; 1 -6 98 1 ; th r ee wa ys str a igh t flow;
LIMIPH ON va lve 1 -9 1 00 ;s tra igh t flow glob e va lve 1 -6 9 32 ;
dou b le s ea t m icr oflow va lve
0 .2 5
a n gle va lve 1 -4 4 11 ; ca ge va lve 1 -4 4 32 ; th r ee wa ys a n gle
va lve; LIMIPH ON va lve 1 -9 4 00
0 .3
12 0 a n gle va lve 1 -4 20 0 ; d ia ph r a gm va lve 1-3 0 0 0; b u tter fly
va lve u p to 4 5 even a t cr itica l flow con dition a n d u p to 9 0
a t su b son ic con d ition .
0 .4
bu tter fly va lve 1-24 71 ; 1 -2 3 1 1 ; 1 -2 5 12 from 4 5 to 9 0 in
critica l flow con d ition - dr illed d isk s
0 .5
Fig. 5 r
w
values for different valve ty pes
Lp(A) 160 10 log Lg sp 10 log T
Wa c r
w
Di
Di 2000
Di
Lfp
2 2
2
+ + + +

+

10
Th e equ a t ion (1) is va lid for s in gle s t a ge va lves . For mu lt is t a ge va lves t h e
s ou n d power is ca lcu la t ed in t h e la s t s t a ge by s u bs t it u t in g p
1
wit h t h e
u ps t r ea m pr es s u r e p
n
.
In equ a t ion (1) a s u pplemen t a r y t er m t a kes in t o a ccou n t t h e a cou s t ic
power gen er a t ed by u ps t r ea m s t a ges .
Fig. 6 Medium values applicable for valve opening 50% and over -
3.2 VALIDITY AND TOLERANCES
Th e equ a t ion (1) is va lid u n der t h e followin g h ypot h es is :
1 . Is ot hropy of t he s ourc e , wh ich mu s t be fr ee t o ir r a dia t e in a n y
dir ect ion .
In ca s e of con t r ol va lves (cylin dr ica l s ou r ce) t h is s it u a t ion in volves a 3
dB n ois e r edu ct ion wh en dou blin g t h e dis t a n ce. Th e pr es en ce of wa lls
clos e t o t h e va lve modifies t h is idea l s it u a t ion by in cr ea s in g t h e s ou n d
level compa r ed t o t h e ca lcu la t ed on e.
For in s t a n ce, wh er e t h e va lve is mou n t ed over a r eflect in g floor, t h e
s ou n d pr es s u r e level is in cr ea s ed by a bou t 3 dB.
2 . Abs e nc e of fore ign dis t urbanc e s
Th e s ou n d pr es s u r e level ca lcu la t ed u s in g t h e equ a t ion (1) is t h e on e
gen er a t ed by t h e va lve. Even t u a l ot h er s ou r ces mu s t be t a ken in t o
a ccou n t by s u it a ble cor r ect ion fa ct or s .
3 . Corre c t i ns t allat i on
Th e va lve mu s t be in s er t ed in t h e pipin g a ccor din g t o s u gges t ion s
ou t lin ed u n der poin t 5.
4 . The t ole ranc e on n ois e es t ima t ion depen ds on t h e va lve t ype wh ich
t h e equ a t ion (1) is u s ed for.
The expect ed t olerance range is t t t t t 5 dB, except for r ot a r y va lves
h a vin g a s oph is t ica t ed des ign , des u per h ea t in g va lves fit t ed wit h in s ide
wa t er in ject ion a n d low n ois e con s t r u ct ion s wit h n ot exa ct ly defin ed
a n d n ot in depen den t pa t h s .
sp corre c t ion fac t or o f spe ct rum
PARCOL
valve t ype
Bu t t erfly valve
1-2471
1-2311
1-2512
Globe valve 1-6911, 1-4411
Cage con ven t ion al valves
1-2473, 1-7251
Cage valve
GBR
LIMIPHON
valve
DN 4" 9,5 3 -5
DN 8" 8,5 2 0
DN 16" 6 -1 +5
11
DESCRIPTION UNITS
NOMENCLATURE
c
2
= Speed of sound in downstream fluid m/ s
Dj = J et diameter mm
Di = Inter nal pipe diameter mm
F
d
= Valve style modifier Dimensionless
fp = Generated peak fr equency Hz
fr = Pipe own fr equency Hz
Lp(A) = A-weighted sound pr essur e level exter nal of pipe dB(A)
Lg = Corr ection for velocity in downstr eam piped dB(A)
M
2
= Mach number in downstr eam pipe =
u
2
c
2
Dimensionless
M
vc
= Mach number at vena contracta at
subsonic condition Dimensionless
Mj = Fr eely expanded jet Mach number Dimensionless
p
1
= Valve inlet absolute pr essur e Pa
p
2
= Va lve ou t let a bs olu t e pr es s u r e Pa
p
2 c
= Va lve ou t let a bs olu t e pr es s u r e a t cr it ica l
flow con dit ion s Pa
p
vc
= Absolute vena contracta pressure at subsonic
flow conditions Pa
p
vcc
= Absolute vena contracta pr essur e at critical
flow conditions Pa
q
m
= Ma s s flow r a t e kg/ s
r
w
= Fraction of acoustic power transmitted
downstr eam Dimensionless
S = Pipe wall thickness mm
T
Lfp
= Acoustic attenuation at peak fr equency dB
T
L
= Acou s t ic a t t en u a t ion dB
s p = Spectrum correction factor dB
u
2
= Average fluid velocity in downstr eam pipe m/ s
u
vc
= Fluid velocity in the vena contracta m/ s
Wa = Acoustic power W
Wm = Str eam power of mass flow W
Wm
2
= Str eam power at valve outlet W
Wm
vc
= Stream power in the vena contracta W
= Acoustic efficiency Dimensionless

2
= Density of fluid at valve outlet kg/ m
3
= Specific heat ratio = cp/ cv Dimensionless
SYMBOL
u
c
2
2
12
Fig. 7 Ty pical Fd values for PARCOL control valves .
More accurate values available on reques t
3.3 ACOUSTIC ATTENUATION
It is u t mos t n eces s a r y t o kn ow t h e a cou s t ic a t t en u a t ion of t h e pipin g t o
pr edict t h e n ois e of con t r ol va lves , ma in ly t o des ign low n ois e on es .
Th a n ks t o con ven ien t a pproxima t ion s , a ca lcu la t ion met h od wa s r ecen t ly
ach ieved [1], s u it a ble for low n ois e va lves ; t h e mos t impor t a n t fea t u r e of
t h is met h od is t h e ch oice of t h e n ois e pe ak fre que nc y a s a n es s en t ia l
va r ia ble for T
L
ca lcu la t ion .
Un der t h e h ypot h es is t h a t n ois e fr equ en cy fp is h igh er t h a n t h e own
pipin g fr equ en cy fr (ma s s a ct ion la w va lidit y) a n d t h a t coin ciden ce
fr equ en cies are lower t h an r es on an ce fr equ en cies , t h e acou s t ic at t en u at ion
T
L
ca n be ca lcu la t ed u s in g t h e followin g equ a t ion :
( )
T
Lfp
10 log 3 10
13
c
2
Di
2
1
2
c
2
415
1
20 log
fp
fr

1
]
1
1
1

S
(2)
wh er e t h e fir s t t er m r epr es en t s T
L
a t fr equ en cy fr a n d t h e s econ d on e t h e
cor r ect ion for pea k fr equ en cy fp.
Th e n ois e pea k fr equ en cy fp ca n a ls o be eva lu a t ed t h eor et ica lly a s a
fu n ct ion of flow con dit ion (s u bs on ic, cr it ica l or h yper cr it ica l) a n d of t r im
geomet r ic s h a pe.
For in s t a n ce, for s u bs on ic flow con dit ion (common in va lves pr ovided
wit h low n ois e t r im) t h e pea k fr equ en cy ca n be ca lcu la t ed u s in g t h e
equ a t ion :
fp 200
u
vc
Dj

(3)
wh er e D
j
is t h e equ iva len t dia met er of t h e jet a t t r im ou t let , wh ich is a
t ypica l con s t r u ct ive da t a of ea ch t r im t ype. It is dir ect ly pr opor t ion a l t o
t h e t r im s h a pe fa ct or Fd, wh os e t ypica l va lu es a r e lis t ed in Fig. 7 t a ble:
Dj 4, 6 10 Fd Cv F
3
L

(4)
Val ve s tyle modifie r Fd
Val ve type
Flow
direct ion
Rela t ive flow
coefficient
0.10 1.00
Globe, pa ra bolic plu g
(1-6911, 1-6951, 1-6921, 1-6981 e 1-4411)
Flow-t o-open
Flow-t o-clos e
0.10
0.20
0.46
1.00
Bu t t erfly valve
1-2471, 1-2512,
1-2311
Ma x. opening
90
60
Wha t ever
0.20
0.20
0.7
0.5
Cage va lve
1-6931, 1-4432,
1-6971, 1-4471
Nu mber of holes
50
100
200
Wha t ever
0.45
0.32
0.22
0.14
0.10
0.07
Dou ble s ea t
1-8110
Par a bolic V-port
Bet ween
s eat s
010
0.10
0.32
0.28
Dj 4,6 10 Fd Cv F
3
L

T
fp
10 log 3 10 c
Di 1
c
415
1
20 log
fp
fr
L
13
2
2
2 2

_
,

1
]
1
1
1
1

S
13
4
LOW NOISE CONTROL VALVES
4.1 DESIGN GUIDELINES
Th eor et ica l pr in ciples for t h e ca lcu la t ion of con t r ol va lve n ois e pr a ct ica lly
defin e t h e des ign gu idelin es of low n ois e s er ies .
Th e a bove ca n be ea s ily s h own by con s ider in g t h e t wo ba s ic pa r a met er s
of con t r ol va lve n ois e:
a cous t i c effi ci ency and p ea k frequency .
4.2 ACOUSTIC EFFICIENCY - MULTISTAGE TRIMS
Fig. 5 s h ows t h a t (for F
L
0.9) t h e r a t io of a cou s t ic efficien cies bet ween
h ypercr it ic flow con dit ion , wit h p
1
/ p
2
>10, a n d t h e s u bs on ic on e, wit h
p
1
/ p
2
= 1.5, ca n a r r ive u p t o 30 ma x, wh ich , a ccor din g t o equ a t ion (1),
cor r es pon ds t o Lp va lu e of 15 dB for t h e n ois e in s ide t h e pipin g.
Th e a bove a cou s t ic a dva n t a ge ca n be t h erefor e a ch ieved on con dit ion
t h a t t h e flu id lea ves t h e t r im u n der s u bs on ic flow con dit ion .
Wh ere h igh pr es s u r e drops mu s t be per for med t h e a bove is on ly pos s ible
by u s in g a t r im pr ovided wit h a s u it a ble n u mber of mu lt iple s t a ges
a r r a n ged in s er ies .
A pr a ct ica l s pecimen of t h is t r im t ype is t h e PARCOL va lve s er ies 1-7251,
s h own in Fig. 8. Th e s pecia l plu g des ign a llows t o s plit t h e pr es s u r e dr op
in t o more s t eps a lon g t h e win din g pa t h cr ea t ed bet ween plu g a n d fixed
s h a ped ou t s ide wa ll.
It is rema r ka ble t h e fa ct t h a t t h e pr es s u r e drop t a kes pla ce t h r ou gh t h e
s in gle s t a ges s imu lt a n eou s ly wit h t h e flow s ect ion a l a r ea r edu ct ion ; t h is
is t h e ba s ic con dit ion for t h e good flow con t r ol qu a lit y.
Th e pr a ct ica l limit s of t h is s olu t ion a r e of con s t r u ct ive n a t u r e a n d ca n be
s u mma r ized a s follows :
1. Ma ximu m n u mber of fea s ible s t a ges
2. Th e expa n s ion r a t io of s ect ion s fr om in let t o ou t let , wh ich in t h e
a for emen t ion ed ca s e s h ou ld be a t lea s t 30:1. As a ma t t er of fa ct it is
n ot s u fficien t t o t a ke ca r e of cr it ica l s t eps wit h ou t min din g t h e flu id
velocit y in s ide t h e t r im.
3. Maximu m r equ ir ed Cv.
14
Fig. 8 Low nois e des ign 1-7251 provided w ith multis tage s ingle path
trim.
Fig. 9 Fixed dow ns tream res trictors
V
15
It is pos s ible t o t r y t o over come t h e fir s t t wo limit a t ion s , ma in ly t h e s econ d
on e, by in s er t in g down s t r ea m h ea d los s es by fixed s ect ion a l a rea t h r ot t les
(s ee Fig. 9).
Th e a bove s u r ely ma kes t h e mu lt is t a ge va lve ea s ier t o con s t r u ct , bu t t h e
pr oces s rangeabilit y get s pr oblemat ic, bot h u nder flow cont r ol and acou s t ic
viewpoin t .
Th is s olu t ion ma y on ly be t a ken in t o con s ider a t ion wh en t h e loa d is r a t h er
con s t a n t a n d a ll of t h e va r ia bles a re kn own ver s u s loa d ch a n ges .
4.3 PEAK FREQUENCY - GBR CAGE TYPE VALVES
Sou n d pr es s u r e levels gen er at ed by con t r ol valves in s ide t h e pipin g almos t
a lwa ys rea ch ver y h igh va lu es .
Lu ckily t h e pipe wa ll a ct s a s a ver y impor t a n t a cou s t ic ba r r ier, wh ich let s
ju s t a s ma ll fr a ct ion of s ou n d in t en s it y pa s s ou t s ide. Ot h er wis e t h e
a cou s t ic pr oblem cer t a in ly cou ld n ot be fa ced n eit h er wit h t h e mos t
s oph is t ica t ed a n d expen s ive low n ois e con t r ol va lves .
As a lrea dy s een u n der poin t 3.3 t h e a cou s t ic a t t en u a t ion of t h e pipe wa ll
is a s s t r on ger a s h igh er is t h e fr equ en cy fp of t h e n ois e compa r ed t o t h e
ma in r es on a n ce fr equ en cy of t h e pipin g.
Th is la w is va lid wh en t h e n ois e fr equ en cy is h igh er t h a n fr, i.e. for h igh
a cou s t ic fr equ en cies (wh ich a re t h e mos t s ign ifica n t u n der t h e a cou s t ic
viewpoin t ) a n d pipe dia met er r ela t ively h igh (low res on a n ce fr equ en cy).
Th en h er e it is a s econ d impor t a n t gu idelin e t o des ign a low n ois e t r im:
Th e a cou s t ic s pect r u m of t h e gen er a t ed n ois e mu s t s h ow h igh er in t en s it y
a t h igh fr equ en cies .
Th e a bove ca n be obt a in ed by kn owin g a ll of t h e a cou s t ic a n d flu odyn a mic
pa r a met er s of t h e ph en omen on , ma in ly of t h e va lve s t yle modifier Fd:
Fd
d
H
do
1
No

(5)
wh er e d
H
an d d
O
a r e r es pect ively t h e h ydr a u lic dia met er a n d t h e on e of
t h e t ot a l equ iva len t flow s ect ion , wh ile No is t h e n u mber of in depen den t
pa t h s a r r a n ged in pa r a llel.
As a lr ea dy s een u n der poin t 3. 3 t h e lea din g fr equ en cy fp is dir ect ly
pr opor t ion a l t o Dj va lu e, i.e. in ver s ely pr opor t ion a l t o Fd.
Hen ce it a ppea r s t h a t , a t a pa r it y of ot h er geomet r ica l va r ia bles , t h e h igh er
is t h e n u mber of pa t h s , t h e h igh er is fp a n d fin a lly t h e lower is t h e n ois e
t r a n s mit t ed t h r ou gh t h e pipe wa ll.
For con ven t ion a l s in gle s t a ge va lves No = 1, except for dou ble s ea t a n d
bu t t er fly ver s ion s , wh er e No = 2.
Acou s t ic b en efit s d er ivin g fr om a cou s t ic a t t en u a t ion a r e t h er efor e
n egligible in t h es e ca s es , s in ce Fd va lu es a r e h igh a n d fp va lu es a r e low.
F d
d
d
o
H 1
No
=
16
Fi g. 1 1 Multicage trim -
The limited number of s tages and paths does not allow to obtain
an acous tic benefit higher than 10 dB.
Fi g. 1 0 GBR ty pe s ingle cage -
The nois e reduction is obtained by providing a very high number
of low diameter holes (24 mm)
Acous tic attenuation up to 15 dB.
17
A low n ois e t r im, bu ilt on t h e ba s is of t h is t h eoret ica l pr in ciple, is t h e
PARCOL GBR model s h own in Fig. 10.
It is a s in gle ca ge model (s in gle-s t a ge, mu lt ipa t h ) provided wit h a ver y
h igh n u mber of s ma ll h oles . Su ch a model a llows t o r ea ch ver y low va lu es
of F
d
(even < 0.02), cor r es pon din g t o fp va lu es h igh er t h a n 20 kHz.
Th e advan t age der ivin g fr om T
L
in creas e mu s t be added t o t h e con t r ibu t ion
of s p, wh ich , du e t o t h e con cen t r a t ion of in t en s it ies a r ou n d fp, n or ma lly
r es u lt s ver y low.
4.4 UNIVERSAL SOLUTIONS MULTISTAGE / MULTIPATH -
LIMIPHON TYPE TRIM
Sin gle pa t h mu lt is t a ge va lve models , like t h e t ype men t ion ed u n der poin t
4.2, t a ke a dva n t a ge fr om t h e low a cou s t ic efficien cy of s u b-s on ic flow
con dit ion , bu t t h eir r ela t ively low pea k frequ en cies limit t h e pipe wa ll
a t t en u a t ion .
Sin gle s t age cage t r ims men t ion ed u n der poin t 4.3 n or mally oper at e u n der
cr it ica l flow con dit ion , bu t t h eir low Fd va lu es a n d con s equ en t ly h igh
fr equ en cies a llow t o pr ofit t h e n ois e a t t en u a t ion du e t o h igh er T
L
valu e.
For bot h t h e a bove ca s es t h e n ois e a t t en u a t ion ca n rea ch 15 dB ma ximu m
(wit h refer en ce t o con ven t ion a l models ), wh ich for s u r e r epres en t s a qu it e
good a cou s t ic per for ma n ce, bu t ma y be on ly obt a in ed wit h a ver y a ccu r a t e
des ign a n d con s t r u ct ion .
Sin ce t h e mos t s ever e a pplica t ion s r equ ir e Lp r edu ct ion s over 20 dB,
mu lt is t a ge/ mu lt ipa s s t r ims wer e s et -u p, t h u s pr ofit in g t h e a dva n t a ges of
t h e t wo a for emen t ion ed s olu t ion s .
A fir s t s t ep t owa r d t h e r ea liza t ion of t h is pr in ciple is r epres en t ed by t h e
mu lt ica ge t r im (Fig. 11), wh ich n ever t h eles s ca n n ot r epres en t t h e t r u e
pr oblem s olu t ion , du e t o s ome t h eor et ica l a n d con s t r u ct ive limit s .
Th e fin a l a n s wer t o t h e mos t s ever e a cou s t ic pr oblems of con t r ol va lves is
r epres en t ed of t h e con t r a r y by t h e PARCOL Limiph on t ype t r im, s h own in
Fig. 12, wh ich is r ea lized by over pos in g met a l dis ks per for a t ed a n d
a r r a n ged a ccor din g t o differ en t pa t t er n s .
No t h eor et ica l limit r ela t ed wit h p
1
/ p
2
r a t io, n u mber of s t a ges a n d s peed
con t r ol exis t for s u ch models .
18
Fi g. 1 2 Trim of LIMIPHON cont rol valves of univers al mult is t age/
multipath ty pe, provided w ith laby rinth dis k s tack .
Fluid paths are obtained by overpos ing dis k s s uitably drilled
and mutually oriented.
Fig. 13 s h ows a t ypica l a pplica t ion of a pr es s u r e r edu cin g va lve of a
met h a n e decompres s ion s t a t ion .
Th e con s t r u ct ion of t h is va lve t ype, yet in t r in s ica lly complex, becomes
ver y exa ct in g wh er e t h e flu id t emper a t u r e is ver y h igh .
Fig. 14 s h ows a HP t u r bin e by-pa s s va lve in t en ded t o r edu ce t h e pr es s u r e
of a bou t 250 t / h s t ea m flow r a t e fr om 100 t o 1.5 ba r ; it s s ou n d pres s u r e
level is 90 dB(A) (ba r e pipe).
Th is va lve t ype is pr ovided wit h a ver y low s pecific Cv t r im a n d gen er a lly
r equ ir es a ver y lon g t r a vel compa r ed t o ot h er models .
19
Fi g. 1 4 Low nois e mod el univers al ty pe s uitable for s ervice on high
t e m p e ra t u re s t e a m - Th e p ict u re s h ow s a v e ry e x a ct in g
application:
by -pas s for condens ation turbine DN 12 x 34
p
1
= 100 bar - p
2
= 1.5 bar - max s team flow rate = 250 t/ h
-max Lp = 90 dB(A) (bare pipe)
Fi g. 1 3 Multis tage/ multipath low nois e ty pe reducing valve, provided
w ith the characteris tic d is k s tack - Mod el s uitable for low
temperature s ervice, lik e s tations for methane gas pres s ure
1s t s tage reduction.
20
5
PIPING INFLUENCE
Nois e predict ion of a con t r ol va lve is a ffect ed by t h e la y-ou t of t h e pipin g
wh er e t h e va lve is in s t a lled.
Redu cer s , elbows , on / off va lves , br a n ch pipes , et c. con t r ibu t e t o gen er a t e
n ois e, like a ll ot h er ca u s es of t u r bu len ce.
Du e t o t h e ext reme pr oblem complexit y it is n ot pos s ible t o ba s e on s imple
cor r ect ion equ a t ion ; ju s t s ome gu ida n ces ca n be given :
- St raight pipe le ngt hs
Min imu m s t r a igh t pipe len gt h s a dja cen t t o t h e va lve n eces s a r y n ot
t o a ffect t h e expect ed s ou n d pr es s u r e level is :
6 DN ups t ream an d 3 DN downs t re am, wh er e DN r epr es en t s t h e
dia met er of t h e body con n ect ion .
Su ch len gt h s in clu d e t h e even t u a l con cen t r ic r ed u cer s wit h
pr ogr es s ively va r ia ble s ect ion s h own in Fig. 16.
Th ey ma y be in cr ea s ed by t h e des ign er a ccor din g t o t h e oper a t ion
h ea vin es s .
- Re duc e rs
To avoid addit ion al n ois e t h ey mu s t h ave a pr ogres s ive s ect ion ch an ge,
ma in ly a t t h e ou t let (s ee Fig. 15). Avoid eccen t r ic fit t in gs .
- On/ off valve s
Wh er e mou n t ed clos e t o t h e con t r ol va lve t h ey s h ou ld be fu ll bor e
t ype (ba ll or ga t e va lves ).
- Elbows , branc he s and ot he r fit t ings
Ea ch s u dden flow devia t ion or flow s ect ion ch a n ges gen er a t e n ois e.
To redu ce t h e acou s t ic in t er fer en ce of s u ch compon en t s it is n eces s ary
t o impr ove t h eir des ign , a s s h own in Fig. 15.
21
Fi g. 1 5 Effect of pipe configuration on s ound pres s ure level of the line
To be avoided
Suitable
Reducers
Curves
Confluences
Branches
Manifolds
Standard curves
Inactive branch
22
6
EXHAUST TO ATMOSPHERE
Th e a cou s t ic pr oblem of t h e dis ch a r ge of a compr es s ible flu id t o t h e
at mos ph ere wh er e t h e n ois e pr opagat es can become ver y cr it ical, becau s e:
1. Th e a cou s t ic in s u la t ion of t h e met a l wa ll is mis s in g
2. p
1
/ p
2
r a t io oft en r ea ch es h igh va lu es , s in ce t h e ba ck pr es s u r e is zer o.
Th is pr oblem a t a gla n ce a ppea r s on ly s olva ble by in s t a llin g a s ilen cer
(expen s ive) on ea ch exh a u s t t o t h e a t mos ph er e.
Lu ckily t h is s olu t ion ca n be oft en a voided for t h e followin g r ea s on s :
- Th e fr ee exh a u s t ca n be con s ider ed a s a pu n ct u a l s ou r ce, wh os e Lp
decr ea s es 6 dB by doubling the dis tance
- Fr ee exh a u s t s a r e n or ma lly lea d t o a cer t a in dis t a n ce from pos s ible
h ea r in g pla ces .
- Fr ee exh a u s t s a r e n or ma lly dis con t in u ou s (s a fet y va lves , s t a r t -u p of
pla n t s , decompr es s ion s t a t ion s , et c.); t h er efor e h igh er s ou n d levels
ar e a llowed for t h em, compa r ed t o t h e on es a llowa ble for con t in u ou s
du t y equ ipmen t s . Th e USA OSHA r egu la t ion , for in s t a n ce, a llows a
ma ximu m level of 115 dB(A) for a n ois e expos u r e of a qu a r t er of h ou r
ea ch eigh t h ou r s .
Compa red t o equ a t ion s u s ed for piped exh a u s t s , in t h is ca s e t h e dis t a n ce
fr om t h e micr oph on e a n d it s a n gle fr om ch imn ey a xis , mu s t be a ls o
a ccou n t ed.
Equ a t ion (1) ca n be u s ed t o pr edict t h e n ois e gen er a t ed by ven t s , by
a s s u min g TLfp = 0.
Th e ou t let from t h e ch imn ey ca n be con s ider ed a s a s ph er ica l s ou r ce
wit h 6 dB decr ea s in g wh en dou blin g t h e dis t a n ce.
However, du e t o it s dir ect ion a l ch a r a ct er is t ic, t h e gen er a t ed n ois e mu s t
be eva lu a t ed a s a fu n ct ion of t h e a n gle bet ween exh a u s t bea m a n d
micr oph on e dir ect ion (s ee Fig. 16).
Her e it is t h e gen er a l equ a t ion of t h e s ou n d pr es s u r e level:
L =109 +10 log f
pA
vent
10
Wa
r
s
2
-
dove :
wh er e:
r = dis t a n ce of t h e micr oph on e fr om t h e ch imn ey t op m
fs = exh a u s t s t yle modifier (s ee Fig. 17 a s a fu n ct ion of ) dB
= a n gu la r devia t ion of t h e micr oph on e degr ees
L = 109 +10 og f
pA 10
Wa
r
vent
s 2
- l
23
Fi g. 1 6 Microphone dis tance
Fi g. 1 7 Exhaus t s ty le modifier - dB
-10
-5
0
5
10
15
0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180

fs
vertic ale
laterale
svasato
vertical
lateral
flared
24
7
ACOUSTIC INSULATION
Th e n ois e gen er a t ed by t h e va lve pr opa ga t es a lon g t h e flu id down s t r ea m
pa t h wit h ou t s ign ifica n t los s .
Acou s t ic in s u la t ion ca n t h er efor e s olve t h e pr oblem on ly in t h e a r ea wh er e
it is r ea lized.
Pipin g en gin eer s oft en min d t he rmo-ins ulat ing laggings (ver y diffu s ed
on s t eam lin es ), wh ich , be ing ins t alle d along t he whole pipe le ngt h,
become in t er es t in g u n der t h e a cou s t ic viewpoin t eit h er.
Fig. 18 s h ows t h r ee t ypica l la ggin g pa t t er n s , wh os e ph on o-in s u la t in g
ca pa cit y is s h own in Fig. 19.
Un for t u n a t ely a cou s t ic in s u la t ion per for ma n ce of s u ch la ggin gs is limit ed
by s ever a l r ea s on s r ela t ed wit h t h eir in s t a lla t ion .
Her e ar e t h e ma in on es :
- a cou s t ic h oles du e t o a ls o r edu ced s u r fa ces n ot la gged
- a cou s t ic br idges bet ween pipe wa ll a n d ou t s ide la ggin g s u r fa ce
- a cou s t ic a n t en n a s con s t it u t ed by br a n ch lin es or h oldin g legs r igidly
con n ect ed wit h t h e pipin g a n d pa s s in g t h r ou gh t h e la ggin g
- loggin gs n ot complet ely s ea led or over la pped
Th es e con s t r u ct ive det a ils n or ma lly do n ot a ffect t h e efficien cy of t h e
t h er ma l in s u la t ion , wh ile r epr es en t a s er iou s in con ven ien ce a s fa r a s t h e
ph on o-in s u la t in g ca pa cit y is con cer n ed.
If a ll t h e a bove is a dded t o t h e n ois e es ca pe fr om t h e u n la gged pa r t s of t h e
va lve (bon n et a n d a ct u a t or ) it ca n be ea s ily u n der s t ood h ow difficu lt is
t h e s olu t ion of t h e va lve a cou s t ic pr oblem by in s u la t ion compa r ed t o ot h er
in du s t r ia l a n d civil a pplica t ion s .
25
Fi g. 1 8 Patterns of phono/thermo-insulating laggings of piping
NOTES
A pattern is the typical thermal insulation
B and C patterns may also be considered as acoustic insulation
Average attenuations shown in the table are valid for a complete lagging,
properly installed and exempt from antennas and acoustic bridges and refer to
spectra with prevailing frequencies ranging from 2000 to 8000 Hz. For a more
accurate estimation as a function of the actual spectrum taken outside the
piping see Fig. 19.
Actual values are practically lower than theoretical ones ( 5 dB(A)).
PATTERN A
mineral wool ( = 80 kg/m
3
)
canvas
piping wall
1 mm thick aluminium sheet
glass wool ( = 50 kg/m
3
)
S
S
/
2
S
S
/
2
S
S
/
2
mineral wool ( = 80 kg/m
3
)
canvas
piping wall
1 mm thick aluminium sheet
glass wool ( = 50 kg/m
3
)
mineral wool ( = 80 kg/m
3
)
canvas
piping wall
1 mm thick aluminium sheet
septum 6 kg/m
2
1
.
5

S
lead plate (6 kg/m
2
)
PATTERN C
PATTERN B
glass wool
( = 50 kg/m
3
)
ATTENUATION - dB (A)
TYPE S = 50 mm S = 100 mm
A 10 14
B 15 19
C 20 23
26
Fi g. 1 9 Acou s t ic a t t en u a t ion of t h e n ois e ou t comin g fr om t h e pipe, a s
a fu n ct ion of la ggin g t ype (s ee Fig. 18) a n d of it s t h ickn es s -
In s u l a t i o n
t h i c k n e s s
F re q u e n c y
k H z
Pa t t e rn
A B C
S = 5 0
0 . 5 3 . 3 6 . 0 8 . 4
1 4 . 3 7 . 7 1 0 . 8
2 5 . 2 9 . 4 1 3 . 2
4 6 . 2 1 1 . 1 1 5 . 6
8 7 . 1 1 2 . 7 1 7 . 9
1 6 8 . 1 1 4 . 5 2 0 . 5
S = 1 0 0
0 . 5 5 . 5 7 . 9 9 . 9
1 7 . 0 1 0 . 2 1 2 . 7
2 8 . 6 1 2 . 5 1 5 . 5
4 1 0 . 1 1 4 . 8 1 8 . 3
8 1 1 . 6 1 6 . 9 2 0 . 9
1 6 1 3 . 3 1 9 . 3 2 3 . 9
ACOUSTIC ATTENUATION - dB
27
2
1
3
4
5
6
7
8
INFLUENCE OF ACOUSTIC
SPECTRUM
High pea k fr equ en cy n ois e is mor e
a t t en u a t ed by pipe wa ll.
Be ca r efu l: t h e a bove is on ly t r u e if
t h e peak fr equ en cy is h igh er t h an t h e
r es on a n ce fr equ en cy of t h e pipin g.
NOISE PROPAGATES THROUGH
DOWNSTREAM PIPING !
At t en u a t ion d u e t o p ip e wa ll is
s t rict ly r elat ed wit h it s t hicknes s and
dia met er.
ANISOTROPY INCREASES THE
NOISE GENERATED BY THE
VALVE!
Pr es en ce of wa lls or ot h er obs t a cles
c l os e t o t h e p i p i n g c a u s e s t h e
a cou s t ic wa ves t o be r eflect ed, t h u s
in cr eas in g t h e s ou n d pr es s u r e level.
MIND THE NOISE GENERATED
BY FLUID FLOW INSIDE THE
PIPING!
High velocit ies a n d s u dden s h a pe
ch a n ges ca n gen er a t e h igh s ou n d
pr es s u r e levels .
ACOUSTIC INSULATION: WHERE
Ac ou s t i c i n s u l a t i on s ol ve s on l y
loca lly t h e n ois e p r ob lem, b ein g
n egligible t h e a t t en u a t ion a lon g t h e
pipe.
ACOUSTIC INSULATION: HOW
Poor in s u la t ion , h oles a n d a cou s t ic
br idges ca n con s ider a bly r edu ce t h e
la ggin g efficien cy.
MIND OTHER NOISE SOURCES
Th e n ois e gen er a t ed by ea ch s ou r ce
s u ms u p wit h t h e n ois e gen er a t ed
by ot h er s ou r ces .
MIND REVERBERATING
ENVIRONMENTS
Wh e n r oom d i m e n s i on s a r e
s m a l l a n d / or a c ou s t i c
a bs or pt ion coefficien t of wa lls is
ver y low t h e ba ckgr ou n d n ois e
ca n r ea ch con s ider a ble va lu es .
EIGHT RULES FOR A GOOD ACOUSTIC DESIGN
28
Bibliography
[ 1 ] Ba u ma n n , H.D. - A Met h od for Pr edict in g Aer odyn a mic Va lve
Nois e Ba s ed on Modified Fr ee J et Nois e Th eor ies .
ASME Pa per 87 - WA/ NCA-7 28, Dicembr e 1987 -
[ 2 ] ISA S75.17 - 1989 - Con t r ol va lve Aer odyn a mic Nois e Pr e-
dict ion -
[ 3 ] Fa ger lu n d , A. C. a n d Ch ow, D. C. , Sou n d Tr a n s mis s ion
Thr ou gh a Cylin dr ica l Pipe Wa ll -
ASME J ou r n a l of En gin eer in g for In du s t r y Vol. 103, No 4,
November 1981, pp. 355-360 -
[ 4 ] Mu r on i Pa olo - Le va lvole di r egola zion e per pr oces s i in du -
s t r ia li - PEG Mila n o 1991
[ 5 ] Mu r on i Pa olo - Le va lvole di r egola zion e a ba s s a r u mor os it
per le cen t r a li t er mich e -
Con vegn o ATI / Mila n o - Novembr e 1994
0596065 Studio Trevisan - Gallarate
1
0
0
0

-

1
1
/
9
6

-

A
C
A

0
5
9
1
PARCOL S.p.A. Via Isonzo, 2
20010 CANEGRATE (MI) - ITALY
C.C.I.A.A. 554316 - Fiscal code & VAT no. (IT) 00688330158
Telephone: +39 0331 413 111 - Fax: +39 0331 404 215
e-mail: [email protected] - http://www.parcol.com