Mechanisms and Scale Effects of Skin Friction Reduction by
Microbubbles
Takahito Takahashi, Akira Kakugawa, Shigeki Nagaya, Tsuyoshi Yanagihara, and Yoshiaki Kodama
Ship Rearch Institutewho is it
平面设计培训课程Studies on microbubbles were carried out, aiming at its application to full-scale ships. In the basic studies using a circulating water tunnel, it was found out that the local void ratio near the solid wall is important for skin friction reduction.
Experiments in combined microbubble-surfactant condtitons were made. The combined effect was not obrved, but surfactants helped to generate smaller bubbles. The comparison of non-surfactant and surfactant experiments ems to indicate that the bubble size does not influence the skin friction reduction effect by microbubbles.
Experiments using a 50m-long flat plate ship were carried out in a towing tank. Bubbles we re injected
at two streamwi locations to find out the effect of the boundary layer thickness. It was found out that t
he boundary layer thickness has little effect and that the distance from the injection point is the most important factor.
1. Introduction
Microbubbles, which are small bubbles injected into the wall turbulent boundary layer, is a device for reducing skin friction acting on a solid body advancing in water. Its skin friction reduction effect reaches up to 80%in tests using a circular water tunnel [1], and therefore it is regarded as a promising device applicable to full scale ships. But at the same time, the energy needed for injecting bubbles at the hull bottom is not small becau large ships have large water depth against which bubbles have to be injected. Therefore it is important to reduce the amount of injected air in order to put microbubbles to practical u [2]. In this project we aim at reducing the amount of injected air by half (in other words increasing the skin friction reduction effect twice), by elucidating and utilizing the mechanism of skin friction reduction by microbubbles. During the past few years we conducted microbubble experiments using a circulating water tunnel, and we have confirmed that the local void ratio clo to the wall has strong correlation with skin friction reduction [3]. In this fiscal year of 2000, we conducted another experiments of microbubbles using the circulating water tunnel, by adding surfactants to water, in order to get combined effects of microbubbles and surfactants, and to increas
e information on the interaction of microbubbles with wall turbulence. The works will be described in the next chapter.
Another important factor to be investigated for applying microbubbles to full-scale ships is the scale effect. Especially, the persistence of the skin friction reduction effect by microbubbles in the downstream direction from the injection point is needed to estimate the overall efficiency of microbubbles in drag reduction. Therefore we carried out experiments by towing a 50m-long flat plate ship at the maximum speed of 7m/c in the 400m-long towing tank. It was found out that the skin friction reduction by microbubbles persists up to the downstream the end of 50m-long ship [4],[5]. In the practical application of microbubbles to a full-scale ship, the design and location of bubble injection to get maximum drag reduction is important. Different locations in the streamwi direction on the hull correspond to different boundary layer thickness, and therefore it is necessary to find out the effect of boundary layer thickness on the skin friction reduction effect by microbubbles. This fiscal year, another experiment using the 50m-long flat plate ship was carried out to investigate the effect stated above,
by injecting bubbles either at the bow, where the boundary layer is very thin, or at the middle, where the boundary layer is well developed. The results will be shown in Chapter 3.
2. Basic experiments using a circulating water tunnel
2.1 Experimental apparatus
The experiments were carried out using the high speed circulation water tunnel shown in Fig. 1[6]. The test ction has the following inner dimensions, the width 100mm, the height 15mm, and the length 3000mm. Fig.2 shows the detail of the air injection chamber. Air was injected into the flow through a plate with many regularly-spaced 1.0mm diameter holes called an array-of-holes plate shown in Fig.3. The width and length of the injection part was 72mm, and the pitch of the holes are 2.5mm in the streamwi direction and 1.875mm in the spanwi direction, resulting in the total of 277 holes. It was t at the position of 1028mm from the inlet of the test ction. The bubbles produced by the array-of-holed plate were about 1mm in diameter. The local skin friction was measured directly using the shear stress nsor shown in Fig.4, who capacity was 2gf and diameter is 10mm. Photographs of the microbubbles were taken using the tup shown in Fig.5. A YAG lar was ud as a light source, who light sheet penetrates the test ction at the position 30mm from the side wall, where a CCD camera is placed outside.
2.2 The relation between void ratio and skin friction reduction
The skin friction was measured at three speeds in three streamwi locations. The results are shown in Fig.6. Modification of the data to compensate the wall effect was applied as described in [3]. The horizontal axis shows the average void ratio in the test ction defined as
w
Q a Q a
Q
a +≡
α (1)
Fig.3 An array-of-holes (AOH)plate. Hole diameter=1mm.
Fig.1 The high-speed circulating water tunnel
defectorFig.2 Air injection chamber
Fig.4 Skin friction nsor φ
Fig.5 Layout of camera and light source
where a Q : volumetric flow rate of air in the test ction
w Q : volumetric flow rate of water in the whole test ction of 100mm ×15mm
At all the three speeds, the skin friction reduction increas as the amount of injected air increas. At the average flow speed V=5m/c, measured values at three locations agree well with each other, and agree wtih the average curve of measurements by Merkle [7], which is shown as
2.08.040
+=−a e C C f f α (1)
where
a α is the average void ratio in the test ction. At V=7m/c, measured values at three locations are
different from each other. At V=10m/c, although the measured values at three locations agree well with each other, they consistently deviate from Merkle's curve.
工商管理学位The local void ratio
w
a a
Q Q Q +was measured using a sution tube system similar to the one ud by Guin
[8]. The measured results, which have been adjusted so that the air volume integrated throughout the channel height agrees with that measured at the injection point, are shown in Fig. 7. By comparing the data with that in Fig.6, it is clear that the local void ratio clo to the wall has strong correlation with the skin friction reduction.
Fig.7 Local void ratio (Array-of-holes plates)
我好想你英文(a) V=5m/s
(b) V=7m/s
Fig.6 Skin friction reduction by microbubbles (Array-of-holes plate)
(a) V=5m/s (b) V=7m/s (c) V=10m/s 0.5
0.60.70.80.9
10
0.020.040.060.08
0.1
0.12
Qa/(Qa+Qw)
C f /C f 0
0.5
0.60.70.80.9
inni10
0.020.040.060.08
0.1
0.12
Qa/(Qa+Qw)
C f /C f 0
0.5
0.60.70.80.910
0.020.040.060.08
0.1
0.12
Qa/(Qa+Qw)
architecture2.3 U of surfactants for combined microbubble-surfactant effects
Surfactants are known to have significant skin friction reduction effect at relatively low speeds. Kawaguchi et al.[9] conducted drag reduction experiment in the test ction of 500mm width and 40mm height using surfactants, where CTAC (Cetyltrimethyl Ammonium Chloride) was ud as surfactant and Salicylic acid was ud as solvent. At CTAC concentration of 50ppm, they obtained skin friction reduction in the flow speed range 0.5m/c to 1.8m/c, with the maximum reduction of 80% at flow speed around 1.3m/c.
In our experiments CTAC was added to water with concentration up to 40ppm and the speed range up to 10m/c. Fig.8 shows skin friction reduction as a function of average void ratio at three differe
nt CTAC concentration including non-CTAC condition, at V=5m/c. It is en that the effect of CTAC on skin friction reduction effect by microbubbles is very small at this flow speed. Fig.9 shows the skin friction reduction effect as a function of flow speed. At the flow speed range lower than 4.0m/c, CTAC ems to have some skin friction reduction effect. Fig. 10 shows comparison of bubbles with or without CTAC condition. It is clearly en that at the CTAC concentration of 40ppm, the bubbles are smaller than tho in the non-CTAC condition. Considering this with the fact that the skin friction reduction effect by microbubbles shown in Fig.9 is not influenced at all by the CTAC concentration, it ems that the skin friction reduction effect by microbubbles is not influenced by the difference in the bubble size.
3. The scale effect of mictobubbles
0.00.20.40.60.81.01.20.00
0.05
0.10
中文在线翻译英语0.15
your song0.20
0.25
Qa/(Qa+Qw)
Cf/Cf0
CTAC 0ppm CTAC 20ppm CTAC 40ppm
Fig.8 Relation between average void ratio
and skin friction ratio. V=5m/c
Fig.9 Relation between skin friction
reduction and flow velocity
Void Ratio Qa/(Qa+Qw)=0.1
0.0
0.20.40.60.81.01.20.0
2.0
考研英语在线学习4.0
6.08.010.0
Tunnel Flow Velocity U, m/c S k i n F r i c t i o n R a t i o C f /C f 0
CTAC 0ppm CTAC 20ppm CTAC 40ppm
Fig.10
Photographs of microbubbles (V=5m/s, Qa/(Qa+Qw)=0.116 )
(a) CTAC 0ppm
(b) CTAC 40ppm
3.1 Experimental tup
In order to obtain data on the streamwi persistence of the skin friction reduction effect of microbubbles over a distance as long as possible, a 50m-long flat plate ship shown in Fig.11 was constructed. Air was injected through array-of-holes plates at two streamwi locations, i.e. one at 3.0m from the front end and the other at 31.0m from the front end. The injection plate had 1mm diameter holes spaced in the same way as that ud in the water tunnel). The size of the injection plates was 500mm wide, 100mm long, 4mm thickness.
In order to reduce the ship's drag, the width was limited to 1.0m, and the water depth was 45mm . The bottom of the ship was flat everywhere. The body of the model ship was made of urethane foam, and the frames were made of aluminum channels. The whole body was constructed by connecting 4m-long blocks. The bottom of the ship had transparent acrylic windows (700mm x 700mm) for obrving microbubbles, and each window was 4 meters apart.
The local skin friction was measured using skin friction nsors S10W-2, who capacity was 2gf, produced by SANKEI ENGINEERING. They were attached to each window. Their locations P1(Position1), P2, P3, P4, P5, and P6 correspond to 3.5m,4.8m,8.8m,31.5m,32.8m,36.8m from the front end of the ship. The relative positions of (P1-P3) to the bow injection point and tho of (P4-P6) to middle injection point were the same. Total resistance was measured using a load cell (capac
ity 500kgf).
3.2 Experimental results
(a) Total drag
The measured total drag of the 50m-long flat plate ship is shown in Fig. 12. The vertical axis shows Ct, the nondimensionalized total drag S V R C t t
22
1
ρ=
,
where S is the wetted surface area. 0Cf is the Schoenherr skin friction curve, an experimental curve that shows the drag of a flat plate with the same area and length. The form factor k was determined as 0.14, bad on the Prohaska's method. The horizontal axis shows the Froude number, i.e., the ship's speed. The fact that 0)1(Cf k + curve agrees well with experiments shows that the wave-making drag component of this ship is small.
Fig.13 shows the reduction of total drag by
Fig.11 50m long flat plate ship for microbubble experiments
1.60E-03
1.80E-03
2.00E-032.20E-032.40E-032.60E-032.80E-03
3.00E-033.20E-030.00
0.10
0.20
0.300.40
Fn
C t ,C f 0m
Fig.12 Total drag of a 50m-long
flat plate ship
