12388 SEPTEMBER 1976 HY9

--------------------------
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

JOURNAL OF THE
HYDRAULICS DIVISION
OBLIQUE IMPINGEMENT OF PLANE
TURBULENT JETS
By Spyridon Bel taos 1

INTRODUCTION

The impingement of turbulent jets on solid surfaces finds application in several

engineering problems, such as jets issuing from hydraulic outlet works, weirs,
vertical take-off aircraft, and various spraying devices. The normal impingement
of turbulent jets on smooth walls is the simplest geometrical configuration and
has been studied extensively in the past (5,6,7,8,9, 10, 11, 13, 15, 19,20,26,29,30).
In practice, however, impingement is often at an angle different from 90°; thus,
it is of interest to study the effects of obliqueness.
Previous studies of the impingement problem have established the existence
of three distinct flow regions. The free jet region extends from the point of
issue of the jet to some distance above the wall. Wall effects that are transmitted
in the form of increased static pressure are negligible and the flow is practically
identical to the well-known free jet (1,23). In the impingement region the static
pressure rises above the ambient and significant pressure gradients are set up
which cause the flow to turn from its initial direction to a direction almost
parallel to the wall. When the turn is accomplished and static pressures drop
again to ambient values, the wall jet region begins. In this region the flow
exhibits many similarities with the well-known wall jet (I 2, I6,22,24,27,28). Ref.
24 is an annotated bibliography on the wall jet and contains a much more
complete list than the one presented herein.
At present there is relatively little experimental and analytical material regarding
oblique impingement. Schauer and Eustis (26) presented an experimental and
analytical study of plane turbulent impinging jets. Their work concentrates on
normal impingement; however, some data on oblique impingement were included .
The analysis was of the integral type leading to rather complex equations. Kamoi
and Tanaka (14) measured turbulence properties near the stagnation point at
various impingement angles . Some data on the wall pressure and shear stress
Note.-Discussion open until February 1, 1977. To extend the closing date one month,
a written request must be filed with the Editor of Technical Publications, ASCE. This
paper is part of the copyrighted Journal of the Hydraulics Division, Proceedings of the
American Society of Civil Engineers, Vol. 102, No. HY9, September, 1976. Manuscript
was submitted for review for possible publication on January 5, 1976.
1 Research Officer, Highway and River Engrg. Div., Alberta Research Council , Edmonton,
Canada.

1177

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1178 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

were included. The oblique impingement of circular jets has been studied by
Yakovlevskii and Krasheninikov (31), Naib (17), and recently by the writer
(4) .
The objectives of the present study are twofold : (I) To investigate flow
properties in the impingement region where a combination of significant pressure
gradients and wall shear stresses cause the severest hydrodynamic action on
the boundary; and (2) to develop a simple predictive method for the wall jet

TABLE 1.-Range of Experiments

Angle of impingement, Nozzle Reynolds

Run number <!>, in degrees H /d number, R 0 •
(1) (2) (3) (4)
I 60 45 .5 12,060
2 60 68.2 12 ,060
3 30 45.5 12,060
4 30 45 .5 8,550
5 30 68 .2 8,550
6 30 68.2 12,060
7 40 45 .5 8,550
8 40 68.2 12,060
9 45 68.2 12,060
10 20 68.2 12,060

FIG. 1.-Definition Sketch

region utilizing existing knowledge regarding wall jet flows . . .

Consider a plane jet impinging at an angle, <!>, as shown in Fig. I. The Jet
issues from a two-dimensional nozzle of width d with an initial velocity, U O •
The height of impingement, H , is defined as the distance from the nozzl~ to
the wall along the jet center line. The wall pressure in excess of the ambient
is denoted simply by p "', and reference to "wall pressure" shall be understood

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1179
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

to mean this pressure excess. The maximum wall pressure, Ps• does not occur
at the point 0, so that the stagnatio n point, S, is removed from Oby a distance,
s.

EXPERIMENTS

The experime nts were conducte d using air as the flow medium. By keeping
air speeds well inside the incompre ssible range, the results should apply to
submerged liquid jets, such as, e.g., free nappes entering stilling basins. Air
was supplied by a compress or connected to a rectangul ar plenum chamber which
terminated in a nozzle smoothly tapered to an exit width of 0.224 cm. The
aspect ratio was about 65, so that the jet could be assumed to be two-dimensional
in its central portion. A 0.635-cm thick plexiglass plate, 15 cm wide and equipped
with 30-cm high sidewalls, was used to produce impingement. This setup is
described in more detail in Ref. 5. Wall pressure was measured by means of
two lines of static pressure holes, 0.16 cm in diameter. The wall shear stress
was obtained using the Preston tube technique (21) and utilizing the calibration
curves of Patel (18).
A total of 10 experime nts were performe d, of which the particulars are
summarized in Table I. The symbols appearing in this table are defined in
Fig. I.

EXPERIMENTAL RESULTS AND ANALYSIS

Impingement Region.- Typical wall pressure distributi ons for <I> = 60° and
30° are shown in Fig. 2. The stagnatio n point, which correspon ds to the location
of maximum wall pressure, does not coincide with the intersecti on of the wall
and the jet center line but is shifted by a distance, s. This shift is caused
by the tendency of the stagnating streamlin e to intersect the wall at right angles.
For the case, <l> = 30°, the pressure is seen to take negative values on the
side of negative x. This occurrenc e has been observed previousl y (14,26) and
is believed to be the result of a vortex formation in the entrainme nt field.
When <l> becomes sufficient ly small, the boundaries of the free jet and the
wall jet get so close ~o each other on the side of negative x (Fig. I) that some
of the fluid in the upper layers of the wall jet is entrained in the free jet,
thus completing a circulator y motion and forming a vortex. At the same time,
the vortex is located sufficient ly close to the wall so as to impress some low
pressures on it.
It is now convenien t to adopt the following conventio n. Let x be the distance
1
along the wall measured from the stagnatio n point, so that the positive x -axis
1
points toward the right-hand side (Fig. I). In order to avoid use of negative
values of x 1 , it is observed that the negative x 1 -axis will be the same with
the positive axis of the suppleme ntary angle 180° - <f>. Thus, by referring flow
properties in the range x 1 < O to the case 180° - <!>, it is possible to dispose
of negative x I values.
For pressure profiles that do not exhibit negative values, let b denote the
value of x I for which p"' = 0.5 p s. Then, if p ..,Ip s is plotted versus TJ = x / b
1
a single curve results, as shown in Fig. 3(a). It follows that p,vf Ps = g(T]) and
the function, g, will be the same as that for<!> = 90°, which is given by (13)

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1180 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

160 RUN I
♦ •oo•
H/ d • H .5
120

,o

"

"0 RUN I
♦ • 30"
120
,,
Hi d• 45 . .S

JOO f
,o i
hs
60
i
,o
i
20

- 20 2,~
0--'--:'c,,__,_-f.J6,---l---!12:-1--,,-i-'-,--'--'0---'---'---'--llnil.o,\~CL.LJL!._j
l-- 1o. ----1
STATI C TAP NUMBER

FIG. 2.-Typical Wall Pressure Distributions

Joi IIUN No. ♦ [ 'I

0 I 60

.□
I 110

110
lO
,o

Pw
P,
• •
0
...,
JO
JO

'
IO
"
20

lb l
0
RUN No
.,.
♦ (' !

..."'. .
150

..
IJ l

FIG. 3.-Wall Pressure Similarity

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1181
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

~ = g(TJ) = e - o.69J,,2 •..••••••••.••••.••••••••••.• (1)

f),
, Fig. 3(a) shows that Eq. I is adequate in the range TJ ::S 1.6,. giving somewhat
low values for larger TJ. This type of similarity has also been reported by Kamoi
and Tanaka (14). However, it ·cannot apply when the profile exhibits negative
values. If pis the minimum pressure (p < 0), then Fig. 3(b) shows that similarity
is preserved in the sense (p.., - p)/(p, - p) = g(TJ). Herein b will be defined
as the value of x 1 , in which p"' - p = 0.5 (p, - p).
Prediction of the wall pressure requires also the knowledge of the quantities,
p,, p, b, ands. Assuming the H/ dis sufficiently large to permit full development
of the jet when entering the impingement region, the jet will be described simply

10 . - - - - - . - - - - . - - - - - ,

8 □ -0.2

Ps H 6
pU}2·~ -0 .4
pU~/2 •7 4 -0.6

2 -0.8
0 i<;__ ____._ _ ____._ _ ____,
- 1.0 L..-~----1-.J........-L..----1--l
0 30 60 90 120 150 180
cp (degrees) cp ( degrees)

FIG. 4.-Stagnation Pressure FIG. 5.-Minimum Pressure

O PRESENT DATA
O SCHAUER & EUST IS
l3 • 8ElTA0 S & RAJ.6.IUTNAM 0.)
E. --0-()'"-;;-...:::..__;:--4---5;0_ ..!..
0

H H
□ b. .6. C A
0 10 0.2
0 PRESENT DATA
A SCHAUER & EUSTIS
0 KAMOI & TANAKA
0 .1
~ ~:l~!:~;1~~~:A:~i~~l~ }♦ 90• s

0 o'-------',o---•J_o_ _ _•Lo_ _ _1J_20---,-'-,o,-----' 0

180 0 )0 60 90
</> (deg,ees) </> (deg,ees)

FIG. 6.-Length Scale for Wall Pressure FIG. 7.-Eccentricity of Stagnation Point
Profiles

by its kinematic momentum flux per unit nozzle width, p ui d. Neglecting the
effects of viscosity (I ,23), the parameters describing the problem will be p,
p ug d, H, and q>. Through dimensional analysis it is easy to show that the
dimensionless groups, [p .,/(p U~/2) ](H/d) , [p/(p Uii/2)] (H/d), b/H, and
s/ H, will be functions of <J>, and thus could easily be evaluated empirically.
These deductions are verified in Figs. 4, 5, 6, and 7 where data for varying
H / d and R 0 are seen to define single curves . Considering the minimum pressure,
p, Fig. 5 indicates that it occurs in the range <l> 2e: 130° and peaks at <l> = 150°.
The stagnation pressure, p., , is evaluated in Fig. 4. Assuming a symmetrical
non-negative pressure profile and equating the pressure integral to the normal
component of the momentum of the jet, Schauer and Eustis (26) derived the
following expression

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1182 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

P, H
- -·- - = 8 sin<!> . (2)
p Ui d
2
This is plotted in Fig. 4. Deviations from the data points correspond closely
to the range of significant values of p, which ·were not accounted for in this
derivation.
Fig. 6 shows the relative length scale, b/ H. The data of Schauer and Eustis
were obtained for H/ d = 20, 30, and 40, while those of Kamai and Tanaka
represent the cases H/ d = 12 and 16. In the range cps 120°, b / H varies little ,
although unexpected behavior may be encountered for cp :,s 20°, in which /J
tends to become undefined.
The eccentricity of the stagnation point is evaluated in Fig. 7. Schauer and
Eustis (26) proposed the equation
s
- = 0. 154 cot¢ . . . . . . . . . . . . . . . . . . . . . . . (3)
H ./

which is seen in Fig. 7 to be adequate at least for cp 2! 30°. This was derived
using potential-flow theory, for an ideal jet issuing at the end of the free-jet
region. However , this derivation made use of the expression giving the location
of the center of pressure, which is consistent with the assumed symmetrical
profile but is not consistent with the observed skew profiles. Further, the width
of the potential jet was assumed equal to that of the real jet at the end of
the free jet region and was evaluated using the expression for cp = 90°, without
accounting for changes in this expression due to obliqueness. Thus, the agreement ·
of Eq. 3 with the data must be regarded as coincidental.
The eccentricity, s, for a potential jet has been given in Ref. 25 and is a
rather complex function of cp. However, in the range 30° s cps 90°, this function
can be approximated very closely by the straight line (3)

:r = 2. 15 ( I - ! )..... .......... ...... ...... .. . (4) ,

in which d = the width of the potential jet. If this jet is placed at the end
of the impi;gement region, its momentum flux will equal p UJ d, if it is required
to have the same momentum flux as that of the real jet at the same location .
Further, instead of matching the respective widths, as was done in Ref. 26.
it is perhaps more realistic to require that the potential jet produces a stagnation
pressure equal to the observed. These requirements lead to

2.15(1 - ~ )
s
...... . . .... ..... . . . . (5)
H f p(<!>)
in which fr(<!>)= [p,/(p, UJ/2)] (H/d). Using the data in Fig. 4, Eq. ~ "':'as
evaluated and is also plotted in Fig. 7 where it is seen to almost comc1de
with Eq. 3. Though the end result with respect to predicting s is prac_tic_ally
the same with that of Ref. 26, the present derivation is free of contrad1ctton,
and thus indicates that there is merit in adopting the potential jet analogy.

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1183
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

Beyond the location of p, the wall pressure increases, gradually approaching

zero. Assuming that in this final stage the profile is similar to that produced
by a i,oint vortex located above the wall, the following relation was obtained
(3), which was found to be an adequate description of the available data
p..,
-_-= [1 +0.1025(Tj -'T])2J-2 . . . . . . . . . . . . . ·. . . . (6)
p

in which 'Tl denotes the locus of p and has an average value of about three
[see also Fig. 3(b)].

1.0
a)
o., 0
RUN No. ip(•)
O 10 20
0.6 0 I 60
• 2 60
A 3 30
o., • 4 30
a 5 JO
■ 6 30
C> 9 45
<J I 120
0.2 V 2 120
D SElTAOS & RAJARATNAM, /ovetoge)

0
To 0 1.0 2.0 3.0 ,.o
Tom
1.0 b)
cog!
'-°~'kl
0.8
,.
0.6

~ RUN No. ♦ 1·1

0.4 a• ~ 3 150
D D 4 150
c,■
a 5 150
0.2 • • 6 150

0
0 1.0 2.0 3.0
1/

FIG. 8.-Wall Shear Stress

0 .08 ~ - ~ -....
,-----,------,-----1r-----i1r------.

•
0.06 0 0
□
0
:cl-o
0 ,.

~
0
0 0
E1 0.04
i-,0 "'::f o
e
PRESENT DATA, RH • 57•103 -IO0•I03
PRESENT DATA (TWO EXP ' TS) (:)
Q_
0 SCHAUER & EUSTIS, RH• 270 • 103 O
0.02 '- D BELTAOS & RAJARATNAM, RH•35 •103 -62•103
o. CORRECTED DATA

0 L---..L...---...l'-------'----.,__1_ _ _.,___ _____,

Q 30 60 90 120 150 180

4> (degrees)

FIG. 9.-Maximum Wall Shear Stress

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1184 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

Considering now the wall shear stree, To , the writer and Rajaratnam (5,6)
have shown that T 0 can be predicted by applying the equation of motion at
the wall and using the similarity of static pressure observed in the impingement
region . Assuming that in the case of oblique impingement, the similarity of
static pressure is preserved, at least very near the wall where the stagnating
streamline is very nearly perpendicular to it, it is possible to derive (3) the
shape of the shear stress distribution
To
- = erf (0.83317) - C 117g(17) . . . . . . . . . . . .. . . . . . . . . . . . . (7)
7 am

in which C 1 = a coefficient depending upon cp; and the maximum wall shear
stress, T 0 ,,,, is

7 om H
d = C.(cp) .. . . . . . . . . .. . . . .. . . .. . . . . . . . . . (8)

The coefficients, C I and C., are to be determined empirically for different

values of cp. Derivation of Eq . 7 assumes a non-negative Gaussian pressure
profile , so that it cannot be reliable for <I> ~ 130°. The available data for cp :S I 30°
is plotted according to the formulation of Eq. 7 in Fig. 8(a) where it appears
to define a single curve in the main portion of the impingement region (17 :S
2.5) . Further, Eq. 7 is shown to give an adequate description of the data using
an average value of 0.2 for C 1 • Fig. 8(b) shows a similar plot for <I> = 150°,
which indicates that Eq. 7 is only a crude approximation to these data.
Considering the maximum shear, Torn• dimensional analysis will show that
T H
__o,_,,__ = C.(<!>, RH) . . . . . . . . . . . .. . . . . . . . . . . . . . . . . (9)
p U~ d

2
=
in which RH RO VR7"d. However, the effect of viscosity on T O appears to
be negligible , as was also found for cp = 90° (13), so that Eq. 9 agrees in essence
with Eq. 8. This is verified in Fig. 9 where data covering a range of RH from
35,000-270,000 define a single curve. In the range <!> ~ 90°, C. varies little ,
while for cp > 90°, it decreases rapidly, and seems to extrapolate to Oat<!> = 180°.
Some difficulties in measuring To for cp > 90°, seem to have introduced errors
in these data (3) . This was detected when it was observed that the gradient
of T O with respect to x I was not the same on both sides of the x 1 -axis near
the stagnation point. On the assumption that the error is , in any one run,
proportional to the measured value of T O , a correction was applied to match
the gradients on both sides of the stagnation point. Corrected data are a lso
shown in Fig. 9.
Finally, due to the presence of considerable pressure gradients in the impinge-
ment region, the Preston tube method is liable to errors larger than 6% , within
a certain distance from the stagnation point. Using the criteria given by Patel,
the critical values of TJ for different <I> values were computed (3) and increase
from 1.9 at <I> = 30° to 2.4 at <I> = 150°.
Wall Jet Region.-In this section, a simple analysis will be presented and

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1185
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

will be compared against the data of other investigators .

Studies of the classical wall jet have established (23) the similarity of velocity
distributions along the vertical direction . If u is the velocity component parallel
to the wall and u,,, its maximum, profiles of u at different values of x reduce
to a single similarity curve, in the following manner
u y
- = f(~); s = -8 . . ...... . ..... .. . . . . . . . . . . (10)
u,,,
in which o. = the larger value of y such that u = 0.5 u . This similarity has
been verified for impinging jets by several investigat~~s for both, normal
(5,13 ,20,26) and oblique (4, 17,26,31) impingement.
Using the similarity function , one could predict the velocity field if the scales,
u III and 13., were known. However, before proceeding with the scales , it is
convenient to study the wall shear stress , To, first.
Assuming that a defect-type of law applies in the inner layer of the wall
jet

U -
--=f1 - 11111 ( Y) ; (11)
u. o
in which u. = ~ ;
and 13 = the value of y whenever 11 occurs . The 111

defect law is valid in boundary layers on smooth and rough walls alike. Schwartz
and Cosart (27) have indirectly verified it for plane jets on smooth walls, while
Rajaratnam (22) found it to hold for rough walls. Applying Eq . 10 at y = 13
gives I= [(13/o.), so that o/o. =constant.Ther efore Eq. II can be rewritten
as
u- 11111 0
--= t,<s) ; s $ - . . . . . . . . . . . . . . . . . . . . . . (12)
11. - o.
Comparing Eqs . 10 and 12 yields

u• . .. . .. . . . ..
t(~) = 1 +-f2<s).. . ... . . . ... . . . . . . (13)
u,,,
which can only be satisfied with a constant, u. / u,,,. It follows that the local
friction factor, Cf, defined as T 0 /(pu~1./2) is independent of x. Further, Eq.
13 shows that Cf depends only upon the shape of the functions, f and [ 1 ,
which are not likely to change with <!> or H/ d but could depend upon the
nozzle Reynolds number, R O • It is therefore postulated that
. . . (14)

A reanalysis of the data of Ref. 26 is presented in Fig. 10, in which u. / u 111

(= VC/2) is plotted versus x/H. These data were obtained with R0 ranging
from 39,000-43,000 . For <!> varying between 30° and 130° and H/ d between
20 and 40, it is seen that u. / u,,, is indeed independent of x/ H, H/ d, and
· <p. Thus, for an average Reynolds number, R0 = 41,000, C 1 = 0.0054. Since
C 1 is postulated to be a function of R 0 only, earlier results for <I>= 90° as
well as for the classical wall jet (<!> = 0°) should apply. Available experimental
results have been reanalyzed by the writer (3) and are plotted in Fig. 11. The

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1186 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

~roken lines indicate average results and extend only in the range of R covered
0
m the respectiv e experiments. The solid line is a reasonable average and
has
the equation
c, = o.0474 R0 11s . . .. . . . . . . . . ... ... . . .. ... .. ... os)

6.0
N
oc~io..<&
• -.A~-- o....,_ --..,,.. ..-
S?
X
4.0
• DATA Of•SCH.6.UER & EUSTIS
·I
:, :,E ♦ 1'1
O 50
Hi d
,o C
♦1'1
70
Hid
30
2.0 o 90 ,o ■ 110 30
b 90 20 ◊ 130 30
• 30 30
0
0 20 4.0 6.0 ,.o
x/H

FIG. 10.-Cons tancy of Friction Factor

o., r----,---- ,-,-,....,.. ,.,,,-----, ---,--,--- ,--,--r-r~ ---~~~

0.08

0.06

Cfllg_Q,4 O SCHAUER & EUSTIS

0 BELTAOS & RAJARATNAM , •~ 90"
SIGAllA, ♦ = o•
MYERS et 01, f ~ O•
6 MYERS et ol, ♦= o•
0 .02 ._,~~_._.,_. __.._._,..__ _,___,.._.L..JL J....L..J..J..L_ _ .____.__._.LJ
2•10 3

FIG. 11.-Eval uation of C1 as Function of R0

1.0 , - - - - - , - - - - - , - - - r - - - - - r - - r - - - - , - - - ~ - - - ,
♦ !'I Hid DATA OF SCHAUER & EUSTIS
0 50 40
0.8

0.6
a.
H
0.4

0.2

0 L..-.--L--L--~---'----'---'---'-
0 2.0 - -'
4.0 6.0 8.0
x/1:i

FIG. 12.-Line arity of 6.

Considering now the velocity field, an analysis of the equations of motion

and continuity based on the property of similarity (Eq. 10) will give (3,23 ,27)
8. = Cx . . (16)
(17)

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1187
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

a = -~
2
(1 + _S_) 2FC
. .. ...... . (18)

in which C, C"', and a= constants; and F = the numerical value of the int~gral,
J;f2d~ = 0.7. The overbar above x denotes that xis measured from a_virtual
origin, not necessarily the same for o. and u"'. The data of Ref. 26 1s seen
to support Eq. 16 in Fig. 12 and is described by the linear relation

~ = 0.089(; + 1.2) ........ ........ ........ .... (19)

For normal impingement, Cartwright and Russell (9) reported C = 0.056 for
R0 in the order of 10 5 . The virtual origin was located at -2.5H. For the classical
wall jet (<!> = 0°), Sigalla (28) gives C = 0.065 for similar R 0 , with the virtual
origin at -0.5d. The results of Schwartz and Cosart (27) for <!> = 0° show C
to decrease from 0.085 at R O = 13,500 to 0.060 at R O = 41,600, with the virtual
origin located between - 7.2d and -15.6d. These consideration s suggest that
C depends somewhat on R 0 and generally it decreases with increasing R 0 •
However, it is not possible at present to delineate the relationship between
C and R 0 , because C depends also upon conditions at the nozzle which vary
among different investigations and possibly are far more significant in causing
C to change. The location of the virtual origin appears to depend only upon
conditions at the nozzle. An overall average equation for o., based on the
preceding works, is

0~ = 0.0765 H (X + 1.35) ; average . . . . . . . . . .. (20)

It remains now to evaluate the velocity scale u,,,. Considering Eq. 18 and
using F = 0.7 and C = 0.0765, the term Cf /2FC, will be less than 0.06 for
R0 2': 20,000 (Fig. 11). In most practical situations, R 0 will be larger than 20,000,
so that a "" -0.5, and
urn = C rn x-O.S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (21)

This approximation is equivalent to neglecting the frictional force, which results

in momentum preservation and thus to a decay of u,,, analogous to that observed
in a free jet. With the same assumptions as before, u,,, will be a function
of p U~ d, p, <!>, H, and x, or after some dimensional analysis, (u,,, / U0 ) V H/ d =
h(x/ H, <!>), which in view of Eq. 21 must be of the form (23):

Um

u, Vd -
{H
-JI'
C 11 (<J>)
QC
Lt,,,
u, - ,f!; ...... ............
C 11 (<J>)
(22)

in which C 11 = a function of q,; and the overbar has been dropped from x
for simplicity. Thus, the preceding approximatio n with respect to the exponent,
a, has led to the conclusion that u"' is independent of H. The data of Ref.
26 have been reanalyzed and are plotted in Fig. 13 in the form (U O/u "' )2 versus
x Id , where the slopes of the resulting straight lines determine corresponding
values of C 11 • These are subsequently plotted in Fig. 14 where they are seen

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1188 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

to define a single curve and thus confirm the preceding analysis. Cartwright
and Russell (9) reported the surprising relation, u a: x- 0 -39 (which would imply
~hat the wall jet is gaining momentum). Howev~~, when this · was reanalyzed
it was found that in the range of x/ d in which their data were obtained, Eq.
22 was valid with C., = 2.41.
The function, C.,(<I>), is now predicted as follows. Let the suffixes l and

60
,1·1 Hid
,o
..•
0 90
50 90 20
50 <O
30 30
,o 0 70 30

0
• 110
130
30
30
2
~~]
(DATA OF SCHAUER & EUSTIS)
[ 30

20

lO

20 ,o 60 60 100 120 IAO 160

x/d

FIG. 13.-Velocity Scale 11,..

4.0
Hid

3.0
• 30}
0 40 REF. 26
I:,. 20
Cu □ REF. 9

2.0

1.0

0
0 30 60 90 120 150 180

cf> (degrees)

FIG. 14.-Evaluation of C,. as Function of <I>

2 denote quantities occurring at the side of positive and negative x, respectively .

The momentum flux, M, of the wall jet is given by f'0pu 2 dy, or taking into
account the previous findings
M = pF C?, C UJ d .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . (23)
Neglecting the frictional forces which were seen earlier to be very small (in
this particular application they also tend to cancel each other) the difference

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1189
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

· ion
M 1 - M 2 should equal the x-proJect · o f the 1·nitial momentum flux of th e
jet, i.e.
pd u~ CF(C~ 1 - c~ 2 ) =Put d cos <t>
or the equivale nt: C~ 1 - C~, 2 = m 2 cos <t> ... • • • · . . . . . . (24)
in which 111 2 = I/ FC. It was found earlier, when considering the eccentriG
~ty
of the stagnation point, that an analogy could be drawn with a correspon
di_ng
jet of potential fluid. Pursuing this line of thought it is assumed that the
ratio,
M1/ M 2 , behaves as that of a potential jet, i.e. (2)
M, c~II I+ cos<t>
-=--= . . . . . . . . . . . (25)
M2 C~ 2 1-cos<t>
Solving Eqs. 24 and 25 and dropping the suffixes and 2 by allowing <t> to
vary between 0° and 180°, gives

C == m
II -yf1 + cos2 <l> . . . . . . . . . . . . . . . . . . . . . . . . . . . . . (26)

For the classical wall jet (<l> = 0°) C,, = 3.32 (16). Substituting in Eq. 26 gives
,n = 3.32, and therefor e

c,, = \! 5.5(1 + cos <l>); 0:S <t> :S 180° . . . . . . . . . . . . . . . . . . . (27)

This is shown to describe the experime ntal results adequate ly in Fig. 14.
Note
that using the average value of 0.0765 obtained earlier for C, would give 111
= 4.3,
which does not agree with the value used in Eq . 27 . The discrepa ncy is attribute
d
to virtual origin effects which were not taken into account in this simple analysis.

SUMMARY AND CONCLUSI ONS

The oblique impingem ent of plane turbulen t jets has been studied in
this
paper. The wall pressure and shear stress distribut ions in the impingem ent
region
have been investiga ted experime ntally and semi-em pirical methods were
devel-
oped for their predictio n. The eccentric ity of the stagnatio n point was
shown
to be in fair agreeme nt with that of a potential jet which has the same
initial
momentum and produces the same stagnatio n pressure as the real jet.
For the wall jet region, an attempt has been made to generalize the conventi
onal
similarity analysis of the classical wall jet. Firstly, it was shown that simultan
eous
validity of similarity of velocity profiles and the defect law in the inner
layer
require the local skin friction to be a constant dependin g only upon
the jet
Reynolds number. This was verified using available experimental results
and
including the classical wall jet as a special ca,se of impingement (impinge
ment
angle = 0°) . An average empirical equation was proposed showing the
local
friction factor to decrease as the I/ 5 power of the Reynolds number,
at least
in the range R 0 = 5,000-300,000.
The jet thicknes s was verified to grow linearly at a slope which is independ
ent
of impingement angle and height , but which seems to depend somewh
at on
the Reynolds number. This relationship is rather difficult to establish because
of significant nozzle effects that vary among differen t investigators

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1190 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

Further, it was shown that for RO 2: 20,000, the velocity scale decays approxi-
mately as the inverse square root of distance which, together with dimensional
reasoning, indicated that u 111 should be independent of the height of impingement.
This was verified using available experimental results . Finally, a satisfactory
theoretical method was developed for predicting the effects of the impingement
angle. This was based on momentum considerations and on an analogy drawn
from the corresponding potential jet.

ACKNOWLEDGMENTS

This work is part of a dissertation submitted to the Department of Civil

Engineering, University of Alberta in partial fullfilment of the requirements
for the degree of Doctor of Philosophy. The writer is grateful to his supervisor ,
N. Rajaratnam for his guidance, constructive criticism, and encouragement
throughout the course of this work. Thanks are due to S. Lovell, who performed
a good portion of the measurements and provided valuable assistance in preparing
the experimental setup. Financial assistance provided by the National Research
Council of Canada is gratefully acknowledged.

APPENDIX 1.-REFERENCES

I. Abramovich, G. N ., The Theory of Turbulent Jets, L. H. Schindel , ed. The MIT

Press, Massachussetts Institute of Technology, Cambridge, Mass., 1%3.
2. Batchelor, G. K., An Introduction to Fluid Dynamics, Cambridge University Press,
Cambridge, England, 1967.
3. Beltaos, S., "Turbulent Impinging Jets," thesis presented to the University of Alberta,
at Edmonton, Alberta, Canada, in 1974, in partial fulfillment of the requirements
for the degree of Doctor of Philosophy.
4. Beltaos, S., "Oblique Impingement of Circular Turbulent Jets," Journal of Hydraulic
Research, International Association for Hydraulic Research, Vol. 14, No. 1, 1976,
pp. 17-36.
5. Beltaos, S., and Rajaratnam, N., "Plane Turbulent Impinging Jets," Journal of
Hydraulic Research, International Association for Hydraulic Research, Vol. 11, No.
I, 1973, pp. 29-59.
6. Beltaos, S., and Rajaratnam , N., "Impinging Circular Turbulent Jets ," Journal of
the Hydraulics Division, ASCE, Vol. 100, No. HYIO, Proc. Paper 10852, Oct., 1974,
pp . 1313-1328.
7. Bradbury , L. J. S., "The Impact of an Axisymmetric Jet onto a Normal Ground ,"
Aeronautical Quarterly, Vol. 23, May, 1972, pp. 141-147.
8. Bradshw, P., and Love, E. M ., "The Normal . Impingement of a Circular Air Jet
on a Flat Surface," British Aeronautical Research Committee Reports and Memoranda
No . 3205, Sept. , 1959, pp. 1-8.
9. Cartwright, W. G., and Russell, P. J., "Characteristics of a Turbulent Slot Jet Impinging
on a Plane Surface," Proceedings, Institute of Mechanical Engineers, Paper 32, Vol.
182, Part 3H, 1%7-1%8, pp. 309-319.
10. Cola, R., "Energy Dissipation of a High Velocity Vertical Jet Entering a Basin,"
Proceedings of the 1/th International Association for Hydraulic Research Congress,
Leningrad, U.S.S.R., Vol. 1, 1%5, pp. 1.52, 1-13. .
11. Gauntner, J. W., Livingood, J. N. B., and Hrycak, P., "Survey of Literature on
Flow Characteristics of a Single Turbulent Jet Impinging on a Flat Plate," TN-D-5652,
U.S . National Aeronautics and Space Administration, Washington, D.C., 1970.
12. Glauert, M. B., "The Wall Jet," Journal of Fluid Mechanics , Vol. 1, Dec., 1956,
p. 625-643. . . . b I J
13. Hrycak, P., et al., "Experimental Flow Charactenst!cs of a Single. Tur u ent et
Impinging on a Flat Plate," TN-D-5690, U.S. National Aeronautics and Space

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 HY9 OBLIQUE IMPINGEMENT 1191
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

Administration , Washington , D.C., 1970.

14. Kamoi, A., and Tanaka, H., "Measurement s of Wall Shea_r Stress, Wa_ll Pressur~
and Fluctuations in the Stagnation Region Produced by Oblique Jet Impingement,
Fluid Dynamic Measurements , Conference Papers, Leicester University Press, Vol.
I , 1972 , pp. 217-227 . . .. .
15. Murray , R. I., Nelson, L. E. and Steven, J. R., (P. G . Hubbard-:-A?v 1ser) , ~fle~t1on
of a Jet by a Normal Wall," in "Seven Exploratory Studies m Hydraulics, by
Hunter Rouse, Journal of the Hydraulics Division, ASCE, Vol. 82, No . HY4, Proc.
Paper 1038, Aug., 1956, pp. l038-26-I038-35.
16. Myers, G. E., Schauer, J. J. , and Eustis, R. H. , "The Plane Turbulent Wall Jet,
Part I-Jet Development and Friction Factor," Teclinical Report No. I, Department
of Mechanical Engineering, Stanford University , Stanford, Calif., June, 1961.
17. Naib, S. K. A.. "Deflexion of a Submerged Round Jet to Increase Lateral Spreading,"
La Houille Blancl1e, Vol. 29, No. 6, Oct., 1974, pp . 455-461.
18. Patel, V . C. , "Calibration of the Preston Tube and Limitations on its Use in Pressure
Gradients," Journal of Fluid Mec/ianics, Vol. 23, 1965, pp . 185-208.
19. Poreh, M., and Cermak , J. E., "Flow Characteristics of a Circular Submerged Jet
Impinging Normally on a Flat Boundary," Proceedings of the 6th Midwestern Conference
on Fluid Mechanics, 1959, pp. 198-212.
20. Poreh, M., Tsuei, Y . G .. and Cermak, J. E. , "Investigation of a Turbulent Radial
Wall Jet," Journal of Applied Mechanics , June, 1967, pp. 457-463.
21. Preston, J. H., "The Determination of Turbulent Skin Friction by Means of Pi tot
Tubes," Journal of the Royal Aeronautical Society. London, England, Vol. 58, Feb.,
1954, pp . l09-121.
, 22. Rajaratnam, N . , "Plane Turbulent Wall Jets on Rough Bounda ries," Water Power,
London, England. April-May-June , 1%7, pp . 149-153 ; 196-201; 240-242, respectively.
23 . Rajaratnam, N . , Turbulent Jets, Elsevier Publishing Co., Amsterdam, The Netherlands,
1976.
24. Rajaratnam, N., and Subramanya, K., "An Annotated Bibliography on Turbulent
Wall Jets ," Technical Report, Department of Civil Engineering, University of Alberta,
Edmonton, Alberta, Canada, 1%7.
25 . Schach , W. , "Umlenkung eines freien Flussigkeitsstra hles an einer ebenen Platte, "
lngenieur Archiv, Vol. 5, 1934, p. 245.
26. Schauer, J. J., and Eustis, R. H ., "The Flow Development and Heat Transfer
Characteristics of Plane Turbulent Impinging Jets," Technical Report 3, Department
of Mechanical Engineering, Stanford University, Stanford, Calif., Oct., 1963.
27 . Schwartz, W. H., and Cosart , W. P., "The Two-Dimensio nal Turbulent Wall Jet,"
Journal of Fluid Mechanics , Vol. IO, June, 1%1, pp. 481-495.
28 . Sigalla, A. "Measuremen ts of Skin Friction in a Plane Turbulent Wall Jet," Journal
Royal Aeronautics Society, London , England, Vol. 62, Dec., 1958, pp. 873-877.
29. Tani , I. and Komatsu, Y., "Impingement of a Round Jet on a Flat Surface, " Henry
Goertler, ed., Proceedings of the /Ith International Congress of Applied Mechanics ,
Munich, Germany, 1964, pp. 672-676.
30. Wolfshtein, M., "Some Solutions of the Plane Turbulent Impinging Jet ," Journal
of Basic Engineering, American Society of Mechanical Engineers, Vol. 92, Dec.,
1970, pp . 915-922.
31. Yakovlevskii, 0 . V., and Krasheninnico v, S. Yu., "Spreading of a Turbulent Jet
Impinging on a Flat Surface," Fluid Dynamics, Vol. 1, No. 4, July-Aug., 1%6, p .
136.

APPENDIX 11.-NOTATION

The following symbols are used in this paper:

a exponent;
b length scale for pressure;
C,C 1 ,C.,C,,.,C,. coefficients;

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192

 1192 SEPTEMBER 1976 HY9
Downloaded from ascelibrary.org by Imperial College of Science, Technology and Medicine on 10/19/23. Copyright ASCE. For personal use only; all rights reserved.

cf skin friction factor;

d = nozzle width;
dv = width of correspond ing potential jet;
F - constant;
f,fp functions;
g Gaussian function;
H inclined impingement height;
h function ;
M = momentum of wall jet;
111 constant;
0 = intersection of jet center line and wall;
fj minimum wall pressure;
P., stagnation pressure;
Pw wall pressure;
RH= Ro VffTJ Reynolds number;
Ro nozzle Reynolds number;
s stagnation point;
s eccentricity of stagnation point;
Uo jet velocity at nozzle;
LI = velocity parallel to wall;
um maximum of LI;
LI . friction velocity;
X = distance along wall, measured from O;
x, = distance along wall, measured from S;
y distance, normal to wall;
6 inner layer thickness;
6. = "one-half" thickness of wall jet;
TJ = x,/b dimensionless distance;
TJ value of 11 denoting location of fj;
V = kinematic viscosity of fluid;
~ = y/6. dimensionless transverse coordinate;
p fluid density;
,. 0 wall shear stress;
Toni maximum of ,- 0 ; and
cl> angle of impingement.

Journal of the Hydraulics Division, 1976, 102(9): 1177-1192