See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/200522549

Aeration Characteristics of Ski Jump Jets

Article  in  Journal of Hydraulic Engineering · January 2008

CITATIONS READS
25 433

4 authors, including:

Lukas Schmocker Michael Pfister

Basler & Hofmann AG University of Applied Sciences and Arts Western Switzerland HES-SO
47 PUBLICATIONS   708 CITATIONS    149 PUBLICATIONS   1,062 CITATIONS   

SEE PROFILE SEE PROFILE

Willi H. Hager
ETH Zurich
343 PUBLICATIONS   6,529 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Extreme Hydrodynamic impact onto buildings View project

The effect of the water body geometry on landslide-tsunamis View project

All content following this page was uploaded by Michael Pfister on 15 April 2014.

The user has requested enhancement of the downloaded file.

 Aeration Characteristics of Ski Jump Jets
Lukas Schmocker1; Michael Pfister2; Willi H. Hager, F.ASCE3; and Hans-Erwin Minor4

Abstract: Scour downstream of ski jumps may be avoided by jet deflection to an area where the energy dissipation is accomplished. The
main purpose of this experimental study was the analysis of the jet air entrainment downstream of a ski jump, both for pure water and
preaerated approach flow conditions. A systematic variation of the Froude number and the flow depth in the approach flow channel
resulted in a range of discharge characteristics, whereas the geometry of the ski jump was maintained for all tests. The air concentration
profile was measured at different locations downstream from the ski jump to evaluate the: 共1兲 jet air concentration distribution; 共2兲 location
of minimum air concentration along the mixture flow jet and development of the minimum and the cross-sectional average air concen-
trations; 共3兲 jet trajectories; and 共4兲 process of air entrainment characteristics and jet disintegration. The results demonstrate the significant
effect of the approach flow Froude number, the approach flow depth, and of preaeration on jet disintegration.
DOI: 10.1061/共ASCE兲0733-9429共2008兲134:1共90兲
CE Database subject headings: Aeration; Energy; Hydraulics; Hydraulic structures; Jets; Scour.

Introduction
Ski jumps as applied in hydraulic structures are a main element to
dissipate energy from high-head dams for relatively large unit
discharges. They may also be used as terminal structures of bot-
tom outlets, but then are often curved in plan view. The following
relates to the straight ski jump and extends previous observations,
based on an experimental study at Versuchsanstalt für Wasserbau,
Hydrologie und Glaziologie 共VAW兲, of the Swiss Federal Institute
of Technology ETH, Zurich.
Although ski jumps were incorporated in many hydraulic
schemes over the past decades, relatively few works on their basic
hydraulic features are available. Most observations are site spe-
cific, such that the design guidelines of ski jumps are currently
incomplete. Most of these hydraulic structures are, therefore,
model tested prior to the final design stage. The loss of a ski jump
or damages caused by scour that were previously not accounted
for may result in disaster. The entire spillway in general—and the
dissipator in particular—require a detailed hydraulic consider-
ation 共Vischer and Hager 1995, 1998; Khatsuria 2005兲.
Recent works on ski jumps include those of Juon and Hager
共2000兲 with a literature review on past hydraulic studies of gen-
eral character, and a preliminary investigation on the plane and
spatial flow patterns of ski jump jets. Heller et al. 共2005兲 consid-
ered the two-dimensional ski jump. They presented information
relating to the pressure distribution on the flip bucket, the take-off
angles of both the lower and the upper jet trajectories, the energy
dissipation from the takeoff to the impact sections, and the chok-
1
VAW, ETH Zurich, CH-8092 Zurich, Switzerland.
2
Ph.D. Student, VAW, ETH Zurich, CH-8092 Zurich, Switzerland.
3
Professor, VAW, ETH Zurich, CH-8092 Zurich, Switzerland.
4
Professor, VAW, ETH Zurich, CH-8092 Zurich, Switzerland.
Note. Discussion open until June 1, 2008. Separate discussions must
be submitted for individual papers. To extend the closing date by one
month, a written request must be filed with the ASCE Managing Editor.
The manuscript for this paper was submitted for review and possible
publication on November 22, 2006; approved on May 25, 2007. This
paper is part of the Journal of Hydraulic Engineering, Vol. 134, No. 1,
January 1, 2008. ©ASCE, ISSN 0733-9429/2008/1-90–97/$25.00.
ing features of the ski jump bucket. It was demonstrated that both
the lower and the upper jet trajectories are of parabolic shape with
the take-off angle different from the bucket angle, depending on
the relative approach flow depth and the corresponding Froude
number. Heller et al. 共2005兲 also investigated scale effects, and
found that water jets issued into the air require a minimum ap-
proach flow depth of the order of 40 mm.
The main purpose of the present research was to analyze the
air entrainment characteristics of a plane jet downstream of a ski
jump, both for pure water and preaerated approach flow condi-
tions. Whereas Heller et al. 共2005兲 employed point gauges to
record the jet trajectories, an air-water concentration probe was
used in the present work. The results of this study, thus, relate to
the development of the air concentration of highly turbulent
water, and air-water jets in the atmosphere, and therefore empha-
size the jet disintegration characteristics 共Bin 1993; Ervine 1998兲
and the resulting plunge pool scour 共Mason 1993; Pagliara et al.
2006兲. The present results may also be relevant to considerations
of water jets for fire fighters, yet with the opposite aims. Whereas
water jets should remain highly compact for fire fighting, ski
jumps should produce an air-water flow with a minimum potential
for plunge pool scour. If the jet thickness is too large, the throw-
ing distance too short, or the impact angle too steep, then plunge
pools may become excessively deep. The addition of special ele-
ments to artificially disperse a water jet issued from a ski jump
was not tested herein.
Hydraulic Model
The hydraulic model of Heller et al. 共2005兲 was used 共Fig. 1兲. The
horizontal approach flow channel was connected to a jet box pro-
viding a rectangular jet of approach 共subscript o兲 flow depth ho
and velocity Vo. Air was added to the preaerated jets from an
in-house pressurized air system, thus providing an approach flow
air concentration Co = QA / 共QA + QW兲, where Q⫽discharge and
subscripts A and W relate to air and water, respectively. The air
concentration profile was measured shortly upstream from the
bucket, resulting in the typical air concentration distribution of
spillways 共Kramer et al. 2006兲. At the jet takeoff from the ski

90 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008

Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org
 above the channel bottom, the tailwater channel, and the instru-
mentation used.
An RBI 共France兲 twin fiber-optical probe was employed to
record local air concentrations C ⬎ 0.1%; the local air-water mix-
ture velocity was recorded for local air concentrations C ⬎ 1%
共Boes and Hager 2003; Kramer et al. 2006兲. Difficulties were
experienced mainly in terms of velocity measurement because the
downstream portion of the twin probe was either in the blackwa-
ter jet reach, or signals were unreliable in the aerated jet reach
because of high jet velocity. Additional problems occurred in the
falling jet portion, because of jet disintegration with a composite
of air pockets instead of a more or less homogeneous air-water
mixture flow. It was also observed that the data correlation for
preaerated jet flow was generally poorer than that of nonaerated
approach flows.
Fig. 1. VAW ski jump model The test program included a total of nine experiments 共Table
1兲. In Tests A to C, ho = 0.045 m and the approach flow Froude
numbers were Fo = Vo / 共gho兲1/2 = 3, 5, and 8, with g as the gravi-
tational acceleration. Test D served to check the transverse jet
flow characteristics. In Tests E to G, preaerated approach flow to
jump bucket, the air concentration distribution was affected by
the ski jump with Co ⬵ 0.20 was investigated. In Tests H and I, the
streamline curvature.
effect of ho on the jet characteristics at Fo = 5 and ho = 0.03 and
A fixed bucket geometry was used during the main tests be-
0.07 m was investigated, with ho in Test I reduced to below the
cause the effects of the bucket radius R and the bucket angle ␤
limit value previously stated to avoid scale effects.
had been analyzed by Heller et al. 共2005兲 共Fig. 2兲. The take-off
A test run was conducted as follows: Once the approach flow
angle was ␤ = 30°, the bucket radius R = 0.40 m, and discharges
conditions were established, i.e., the parameters ho, Fo, and Co
were up to Q = 150 L / s in a channel 0.50 m wide. To avoid scale
were set, both air concentration and mixture flow velocity were
effects mainly due to viscosity, the minimum approach flow depth
recorded at each section, from the coordinate origin x = 0 at the
was ho = 0.04 m. Fig. 1 shows the test installation with 共from left兲
take-off section to slightly upstream from jet impact onto the
the approach flow pipe of 250 mm internal diameter, the jet box
tailwater channel bottom. The horizontal spacing of the vertical
allowing for a variable ho, the spillway takeoff elevation 0.304 m
sections was between 0.05 and 0.30 m, and the vertical spacing
5 mm. Depending on ho and Fo, each test included 10 to 15 sec-
tions, resulting in a total of some 300 observational points.
The hypothesis according to which a jet issuing from a ski
jump may be considered nearly plane was investigated with Test
D. Three sections at streamwise locations x = 0.172, 0.492, and
0.812 m were considered, and air concentration and velocity pro-
files measured at transverse locations from the side wall y
= 0.250, 0.125, 0.050, and 0.015 m. The profiles were used to
determine the cross-sectional average 共subscript a兲 air concentra-
tion Ca. Fig. 3共a兲 shows Ca共x兲 and indicates small differences
among the three sections, at least in the near field. One may,
therefore, approximate the axial jet conditions representative for
the entire cross section. The following describes the test results
relating to the axial air concentration and velocity distributions,
Fig. 2. Sketch of ski jump flow investigated, with notation
based on the test program. Results include the concentration pro-
files, contour plots of air concentration, the jet trajectories, a gen-

Table 1. Test Program and Basic Test Parameters

Co ho Q Fo Wo Ro ⫻ 10−5
Test Symbol 关-兴 关m兴 关L/s兴 关-兴 关-兴 关-兴
A ⽧ nonaerated 0.045 74.8 5.0 82 1.3
B 䉱 nonaerated 0.045 44.7 3.0 49 0.8
C 䊏 nonaerated 0.045 119.2 8.0 131 2.1
D 共 a兲 nonaerated 0.045 74.8 5.0 82 1.3
E 〫 0.21 0.045 74.7 5.0 82 1.3
F 䉭 0.19 0.044 44.7 3.1 50 0.8
G 䊐 0.20 0.044 119.0 8.2 132 2.1
H 쎲 nonaerated 0.070 147.8 5.1 130 2.6
I 䉲 nonaerated 0.030 40.0 4.9 54 0.7
a
Particular test for transverse jet effect, but otherwise not accounted for in the data evaluation.

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008 / 91

Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org
 Fig. 4. Side view of jets with Fo = 8, ho = 0.045 m, and 共a兲 Co = 0; 共b兲
Co = 0.20. Compare with Fig. 5 for jet air concentrations.

Contour Plots
The air concentration profiles are shown as contour plots of jets
horizontally stretched. The jet geometry will be investigated
below. Fig. 5 relates to Test C with Co = 0 and Test G with Co
= 0.20, showing the streamwise development of the jet air content.
Fig. 3. 共a兲 Effect of transverse coordinate y on average air concen-
Two particular angles of disintegrating jets were considered in the
tration distribution Ca共x兲, Test D; 共b兲 axial air concentration distribu-
past; namely, the jet core angle ␥i as the limit of the jet core, and
tion C versus dimensionless vertical coordinate Z = 共z − zU兲 / 共zO − zU兲
the transverse jet growth angle ␥ j describing the jet surface where
for Test A
the air concentration has the standard value C = 0.90 共Vischer and
Hager 1998兲. The jet core angle was defined by isolines of jet air
eralized air concentration profile, and the decay of the maximum concentrations of C = 0.10 and 0.20.
streamwise velocity. Fig. 6共a兲 shows a definition sketch for a jet issued from a
nozzle of height ho into a stagnant fluid, in contrast to the present
free surface jet flow. The jet thickness at a specific location is
h j = zO − zU, whereas the core thickness is hi. Both angles ␥i and ␥ j
Test Results
are known not to vary with Fo for turbulent flow. Fig. 6共b兲 shows
h j / ho versus x / 关ho共1 − Co兲兴, and the trend line with a coefficient of
Concentration Profiles determination r2 = 0.88
Fig. 3共b兲 shows the air concentration profiles C共Z兲 at three loca- h j/ho = 1.26 + 0.051兵x/关ho共1 − Co兲兴其 for 共x/ho兲 ⬍ 60 共1兲
tions downstream from the take-off point x = 0 for Test A, with
Z = 共z − zU兲 / 共zO − zU兲 as the dimensionless vertical coordinate rela- From Eq. 共1兲, the jet spread angle is ␥ j = 共1 / 2兲arctan关0.051/
tive to the upper 共subscript O兲 and the lower 共subscript U兲 jet 共1 − Co兲兴, i.e., ␥ j = 1.5 deg, corresponding to a full spread angle of
trajectories. Close to takeoff 共x = 0.092 m兲, the profile is nearly 2␥ j = 2.9 deg for Co = 0 and 2␥ j = 3.6 deg for Co = 0.20, respec-
rectangular, with a blackwater core spanning over 70% of the jet
共subscript j兲 thickness h j. Further downstream, air is entrained
along the upper and the lower jet trajectories by turbulence pro-
duction along the chute and by interfacial mixing along the air-
water boundary layer. As noted from Fig. 3共b兲, the concentration
profiles are not symmetrical about the jet centerline: The location
zm of the minimum 共subscript m兲 air concentration is above
0.50⫻ Z. Whereas water drops ejected from the upper jet trajec-
tory eventually return onto it, those ejected along the lower jet
trajectory fall onto the channel bottom.
Of particular relevance are the upper and the lower jet trajec-
tories zO共x兲 and zU共x兲 plus the location zm and the amount Cm of
minimum air concentration. A definition sketch is shown in the
inset of Fig. 2. Figs. 4共a and b兲 show the jet dispersion effect for
Tests C and G. Upstream from the bucket, the flow is slightly
aerated at the free surface for Co = 0, followed by a dark gray jet
core from the take-off section to almost the maximum jet eleva-
tion, and further downstream, a jet containing a considerable
amount of air finally impacting the channel bottom as a fully
disintegrated air-water flow. For Co = 0.20, the jet thickness is
larger and the impact location closer to the take-off section be- Fig. 5. Contour plot of the streamwise jet air concentration C共x ; z兲
cause of increased turbulence production and energy dissipation. for 共a兲 Test C with Co = 0 and 共b兲 Test G with Co = 0.20, Fo = 8

92 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008

Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org

Fig. 6. 共a兲 Definition sketch of nozzle-issued one-phase flow jet; 共b兲
jet thickness h j / ho versus x / 关ho共1 − Co兲兴 with 共−兲 Eq. 共1兲, jet core
thickness hi / ho versus x / 关ho共1 − Co兲兴 for C = 共c兲 0.10 and 共d兲 0.20
with 共−兲 Eq. 共2兲. Symbols, see Table 1.
JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008 / 93
Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org
Fig. 7. Comparison of jet trajectories according to 共−
− 兲 Eq. 共3兲 with
共−兲 RBI probe 共present study兲 for 共a兲 Test A with Fo = 5, ho
= 0.045 m, and Co = 0; 共b兲 Test G with Fo = 8.2, ho = 0.044 m, and
Co = 0.20
tively. Similar values are reported for turbulent jets 共Rajaratnam
1976; Chanson 1991; Khatsuria 2005; Annandale 2006兲.
Figs. 6共c and d兲 show the jet core thickness hi/ho again versus
x / 关ho共1 − Co兲兴 along the isolines of C = 0.10 and C = 0.20, respec-
tively. Note that the two air concentrations C considered for the
jet core development result in a similar finding, except for the
exact numerical values. The two jet core decay angles are ␥i
共C = 0.10兲 = 1.6 deg, ␥i共C = 0.20兲 = 1.2 deg for Co = 0, and ␥i
共C = 0.10兲 = 2.0 deg, ␥i共C = 0.20兲 = 1.5 deg for Co = 0.20, respec-
tively. The angles for preaerated flows are larger than those of the
nonaerated flows. As for other quantities, Test H has a different
trend in the far jet field. The data plotted in Figs. 6共c and d兲 do not
start at 共0;1兲, due to jet curvature across the ski jump and a slight
surface aeration of the approach flow. Their trend lines may be
expressed with the jet core coefficient D = 0.056, r2 = 0.88 for C
= 0.10; and D = 0.041, r2 = 0.90 for C = 0.20 as
hi/ho = 1.05共1 − D ⫻ 兵x/关ho共1 − Co兲兴其兲 共2兲
Jet Trajectories
The jet trajectories zO and zU of a high-speed air-water flow are
usually defined by a jet air content C = 0.90. The present results
based on the air concentration probe were compared with those of
Heller et al. 共2005兲, in which the jet surfaces were measured with
a trajectory point gauge. Fig. 7 relates to 共a兲 a nonaerated and 共b兲
a preaerated jet. The data of the two samples essentially collapse,
mainly in the near field up to the trajectory maxima. Further
downstream, h j共x兲 is larger in the present study as of Heller et al.
共2005兲. Preaeration of a jet, therefore, increases the jet thickness,
mainly along the upper jet trajectory. This is again due to the
increased turbulence and air addition. The equation of the jet
trajectory is 共Heller et al. 2005兲
Fig. 8. Cross-sectional average air concentration Ca共x兲 for 共a兲 Co
= 0 and various Fo; 共b兲 comparison between Co = 0 and 0.21, for Fo
= 5 and ho = 0.045 m. Symbols, see Table 1.

gx2
z共x兲 = zo + tan共␣兲x − 共3兲
2V2o cos2共␣兲
in which zo = 0 for the lower and zo = ho for the upper jet trajectory,
respectively; and ␣⫽jet take-off angle. Note that Eq. 共3兲 corre-
sponds to the standard parabolic jet profile in which the take-off
angle accounts for the approach flow effects. The present research
supports the previous findings.

Air Entrainment Characteristics

Fig. 8共a兲 shows the streamwise development of cross-sectional
average air concentrations Ca共x兲

冕
zO
1
Ca = C共z兲 ⫻ dz 共4兲
hj zU

All data are seen to follow a trend line, with an initially steep
increase reducing for larger values of x to finally tend toward the
asymptotic value Ca → 1. Fig. 8共b兲 compares the data of the two
Tests A and E with Co = 0 and 0.21 for Fo = 5. The difference
⌬Ca共x兲 between the two remains almost constant along the jet
with ⌬Ca ⬵ 0.15⬍ Co = 0.21 because of air detrainment in the ski
jump reach. Note that Ca共0兲 ⬎ 0 because of the slight surface air
entrainment prior to jet takeoff.
The effect of the approach flow depth ho on Ca共x兲 may be
accounted for by the dimensionless coordinate x/ho. Further, the
relative bucket curvature ho/R and the approach Froude number
Fo have an effect 共Heller et al. 2005兲. The governing dimension-
less coordinate was identified as X = 共x / ho兲共R / ho兲共1 / Fo兲 = x
⫻ R / 共Fo ⫻ h2o兲. The approach flow Reynolds and Weber numbers
are confined to 7.1⫻ 104 ⱕ Ro ⱕ 2.6⫻ 105 and 49ⱕ Wo ⱕ 132, re-
spectively. Here, Ro = Vo ⫻ ho / ␯ and Wo = 共␳W ⫻ V2o ⫻ ho / ␴兲1/2 with
␯ as the kinematic viscosity, ␳W as fluid density, and ␴ as the
surface tension. Excluding Test I from the data set because of
ho = 0.030 m, and with the normalized average air concentration
Fig. 9. Plots relating to 共a兲 ca共X兲, all tests with 共−兲 Eq. 共5兲; Cm共X兲 for
共b兲 nonaerated tests and 共c兲 preaerated tests with 共−兲 Eq. 共6兲, 共d兲
cross-sectional maximum relative velocity V M / Vo共X兲 for preaerated
tests with 共−兲 Eq. 共7兲. Symbols, see Table 1.
ca = 共Ca − Co兲 / 共1 − Co兲 where Co⫽average approach flow air con-
centration, the relation for ca共X兲 may be expressed with r2 = 0.89
as 关Fig. 9共a兲兴
ca = tanh共0.020 ⫻ X兲 for 0 ⬍ X ⬍ 60 共5兲
For small X values, the tangent hyperbolic function in Eq. 共5兲
degenerates to ca = 0.020⫻ X, stating that ca increases linearly
with distance x, the bucket radius R, and decreases with increas-
ing Froude number Fo and the square of the approach flow depth
ho. With Fo = 共hc / ho兲3/2, one has also X = x ⫻ R / 共ho ⫻ h3c 兲1/2 such
that the dominant effect in ca共X兲 is the critical flow depth hc
= 关Q2 / 共gb2兲兴1/3.

94 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008

Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org
 The minimum 共subscript m兲 cross-sectional air concentration
Cm increases with x. The test data were again plotted as Cm共X兲 in
Figs. 9共b and c兲. Except for Test C, all data collapse on a single
curve stating in the near field with r2 = 0.98 共nonaerated兲 and r2
= 0.95 共preaerated兲
Cm = 关tanh共X/30P兲兴3 for 0 ⬍ X ⬍ 60
The data for the preaerated flow with Co ⬎ 0 follow the trend of
those with Co = 0, with the preaeration coefficient P = 2 for
nonaerated, and P = 1 for preaerated flow. Eq. 共6兲 reduces to Cm
= 共X / 30P兲3 for small X/P. Accordingly, the minimum air concen-
tration Cm of preaerated flows with Co ⬵ 0.20 is by a factor of 8
larger than of nonaerated flows. The reason why Test C with Fo
= 8 and Co = 0 deviates from Eq. 共6兲 is unknown.
The velocity data were determined in air-water flows provided
C ⬎ 1%, for which the probe correlation exceeded 0.90. Because
the horizontal velocity component ␯ was measured, the absolute
velocity V was V = v / cos ␦ with ␦ as the local streamline angle
computed from the jet profile Eq. 共3兲. Fig. 9共d兲 shows the relative
maximum 共subscript M兲 cross-sectional velocity of preaerated
jets as V M 共x兲/Vo indicating a significant decay of velocity along
the jet trajectory. The test data may be expressed with r2 = 0.85 as
V M /Vo = 1.12共1 − 0.0125 ⫻ X兲 for 0 ⬍ X ⬍ 50
Additional tests are required to detail this relationship, given the
complexities with spray flow and the limitations of the instrumen-
tation.
Generalized Air Concentration Profile
As mentioned, a concentration profile is made up of the upper and
the lower jet trajectories zU共x兲 and zO共x兲, plus the location zm共x兲
and the value Cm 共Fig. 2兲. A data analysis indicated that both the
upper and lower concentration portions may be represented by
two power functions. The data were normalized as c = 共C
− Cm兲 / 共0.90− Cm兲 providing boundary values c共C = Cm兲 = 0 and
c共C = 0.90兲 = 1; and Z = 共z − zU兲 / 共zO − zU兲 with 0 ⱕ Z ⱕ 1. Constant
values of Zm = 0.63 共±0.05兲 for nonaerated and Zm = 0.54 共±0.07兲
for preaerated flows were used. The data were expressed with a
power function as
c = ␨1/m
in which 共1 / m兲 is an exponent, ␨ = 关共Z − Zm兲 / 共1 − Zm兲兴 for the
upper and ␨ = 关共Z − Zm兲 / 共−Zm兲兴 for lower jet portions, respectively.
The trend of the data follows with m = 0.176 for Zm ⱕ Z ⱕ 1,
whereas m = 0.224 for 0 ⱕ Z ⱕ Zm. Fig. 10 compares the data with
Eq. 共8兲 at dimensionless location X; the data quality of nonaerated
tests is higher than that of the preaerated flows. In addition, the
data scatter appears to increase with Fo, given the highly turbulent
flow pattern. The data for small distances X tend to a more trian-
gular and those for large X to a more rectangular shape than
predicted from Eq. 共8兲. More data under a wider test program
would have to be collected for a final data analysis.
Discussion of Results
The previous results allow for an appreciation of the jet air con-
centration prior to jet impact. If the impact 共subscript s兲 elevation
Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org
is assumed to be identical to the jet take-off elevation, the hori-
zontal distance between the two is ls = sin共2␣兲 ⫻ 关V2o / g兴 from Eq.
共3兲. Further, the relative bucket radius under design conditions is
of the order R / ho = 10 共Vischer and Hager 1998兲. Therefore, the
corresponding dimensionless distance is Ls = 共ls / ho兲共R / ho兲F−1 o
= 10 sin共2␣兲Fo. The average impact air concentration follows
共6兲 from Eq. 共5兲 as Cas = Cas共Co , Fo , ␣兲. The minimum value of Cas
results for Co = 0. Assuming a typical take-off angle ␣ = 30 deg
results in a relation between Cas and Fo, with Cas共Fo = 4兲 = 0.60,
Cas共Fo = 6兲 = 0.78, and Cas共Fo = 8兲 = 0.88. For a jet with Co = 0.20
results in some 10% higher values. Accordingly, for a prototype
condition with a typical approach flow Froude number in excess
of 5, the average impact jet air concentration is always larger than
50%. For small ␣ values, the air entrainment is considerably re-
duced. The take-off angle is the relevant design variable to influ-
ence the jet disintegration process, thereby accounting for the
results of Heller et al. 共2005兲.
A similar analysis may be performed for the minimum jet air
concentration Cms at jet impact, assuming the same basic param-
eters as previously for the average jet air concentration results in
Cms共Fo = 4兲 = 0.14, Cms共Fo = 6兲 = 0.34, and Cms共Fo = 8兲 = 0.55 from
Eq. 共6兲. For preaerated flow, these numbers are significantly in-
creased. Therefore, typical ski jump jets hardly have a blackwater
core, and they are considerably aerated at the impact location. The
共7兲 procedure used, e.g., by Pagliara et al. 共2006兲, for scour hole
experimentation with an air-water mixture pipe flow may be con-
sidered realistic, therefore.
Conclusions
This research accounts for the two-phase flow pattern of plane jet
flow related to ski jumps. The research efforts were focused on
the streamwise development of both nonaerated and preaerated
jets issued by a circular-shaped bucket. The test program included
a range of approach flow Froude numbers, and a special series
with a small approach flow depth to investigate scale effects.
The concentration profiles along a ski jump jet may be de-
scribed using the cross-sectional minimum air concentration as
given in Eq. 共6兲, and the jet trajectories where the air concentra-
tion is 90%. A generalized air concentration distribution is pro-
vided by Eq. 共8兲, thereby accounting for the previous information.
Further, contour plots were discussed in which the complex fea-
共8兲
tures of a turbulent free jet subjected to gravitational forces are
observed. The jet spread and jet core angles were also analyzed.
The average air concentration along a jet increases linearly in the
jet near field and tends to an asymptote in the jet far field. A linear
decay of the cross-sectional velocity was noted for preaerated jet
flow. The minimum and average jet air concentrations at jet im-
pact were also discussed. It was found that the average air con-
centration for typical design parameters is then in excess of 50%.
These indications may be useful for the analysis of highly turbu-
lent air-water jet flows subjected to gravitational forces, and thus
deviating from the standard jets found in engineering applica-
tions. The present results may also be of interest in the analysis of
scour holes associated with a ski jump.
Acknowledgments
The second writer was supported by the Swiss National Science
Foundation, Grant No. 200021-101548/1.
JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008 / 95
 Fig. 10. Generalized air concentration profile c共␨兲 with 共−兲 Eq. 共8兲

96 / JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008

Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org
Notation i ⫽ jet core;
j ⫽ jet surface;
The following symbols are used in this paper: M ⫽ maximum;
C ⫽ air concentration; m ⫽ minimum;
c ⫽ normalized air concentration, c = 共C − Co兲 / 共1 − Co兲; O ⫽ upper jet trajectory;
D ⫽ jet core coefficient; o ⫽ approach flow;
F ⫽ Froude number, F = V / 共gh兲1/2; s ⫽ jet impact location;
g ⫽ acceleration due to gravity; U ⫽ lower jet trajectory; and
h ⫽ flow depth, jet thickness; W ⫽ water.
L ⫽ dimensionless streamwise distance;
l ⫽ streamwise jet distance to jet impact;
m ⫽ exponent; References
P ⫽ preaeration coefficient;
Q ⫽ discharge; Annandale, G. W. 共2006兲. Scour technology, McGraw-Hill, New York.
R ⫽ Reynolds number, R = V ⫻ h / ␯; Bin, A. K. 共1993兲. “Gas entrainment by plunging liquid jets.” Chem. Eng.
R ⫽ bucket radius; Sci., 48共21兲, 3585–3630.
r2 ⫽ coefficient of determination; Boes, R. M., and Hager, W. H. 共2003兲. “Two-phase flow characteristics of
V ⫽ velocity; stepped spillways.” J. Hydraul. Eng., 129共9兲, 661–670.
W ⫽ Weber number, W = 共␳W ⫻ V2 ⫻ h / ␴兲1/2; Chanson, H. 共1991兲. “Aeration of a free jet above a spillway.” J. Hydraul.
Res., 29共5兲, 655–667.
X ⫽ dimensionless streamwise coordinate,
Ervine, D. A. 共1998兲. “Air entrainment in hydraulic structures: A review.”
X = x ⫻ R / 共Fo ⫻ h2o兲; Proc. Inst. Civ. Eng., Waters. Maritime Energ., 130, 142–153.
x ⫽ streamwise coordinate; Heller, V., Hager, W. H., and Minor, H.-E. 共2005兲. “Ski jump hydraulics.”
y ⫽ transverse coordinate; J. Hydraul. Eng., 131共5兲, 347–355.
Z ⫽ dimensionless vertical coordinate, Juon, R., and Hager, W. H. 共2000兲. “Flip bucket without and with deflec-
Z = 共z − zU兲 / 共zO − zU兲; tors.” J. Hydraul. Eng., 126共11兲, 837–845.
z ⫽ vertical coordinate; Khatsuria, R. M. 共2005兲. Hydraulics of spillways and energy dissipators,
␣ ⫽ take-off angle; Dekker, New York.
␤ ⫽ bucket angle; Kramer, K., Hager, W. H., and Minor, H.-E. 共2006兲. “Development of air
␥ ⫽ jet angle; concentration on chute spillways.” J. Hydraul. Eng., 132共9兲, 908–
␦ ⫽ local streamline angle; 915.
Mason, P. J. 共1993兲. “Practical guidelines for the design of flip buckets
␨ ⫽ vertical coordinate relative to Z;
and plunge pools.” Int. Water Power Dam Constr., 45共9兲, 40–45.
␳ ⫽ density;
Pagliara, S., Hager, W. H., and Minor, H.-E. 共2006兲. “Hydraulics of
␴ ⫽ surface tension; and plunge pool scour.” J. Hydraul. Eng., 132共5兲, 450–461.
␷ ⫽ kinematic viscosity. Rajaratnam, N. 共1976兲. Turbulent jets, Elsevier, Amsterdam, The Nether-
Subscripts lands.
Vischer, D. L., and Hager, W. H. 共1995兲. Energy dissipators, Balkema,
A ⫽ air; Rotterdam, The Netherlands.
a ⫽ average; Vischer, D. L., and Hager, W. H. 共1998兲. Dam hydraulics, Wiley, Chich-
c ⫽ critical; ester, U.K.

JOURNAL OF HYDRAULIC ENGINEERING © ASCE / JANUARY 2008 / 97

View publication stats Downloaded 08 Nov 2011 to 129.132.202.35. Redistribution subject to ASCE license or copyright. Visit http://www.ascelibrary.org