International Journal of Engineering & Technology, 7 (3.

29) (2018) 236-242

International Journal of Engineering & Technology

Website: www.sciencepubco.com/index.php/IJET

Research paper

Investigation of base flow for an axisymmetric suddenly

expanded nozzle with micro JET
Zakir Ilahi Chaudhary 1 *, Vilas B. Shinde 2, S. A. Khan 3
1 Dept.
of Automobile Engineering Dept., M H S S College of Engineering, Mumbai, India
2 NHITM, Department of Mechanical Engineering, DMCE, Airoli, Navi Mumbai, Maharashtra, India
3 Mechanical Engineering Dept., Faculty of Engineering, IIUM, Kuala Lumpur, Malaysia

*Corresponding author E-mail: sakhan@iium.edu.my

Abstract

This investigation presents the outcome of the tests conducted to control the pressure in the re-circulation zone. Also, the efficiency of the
flow controllers to govern the pressure at the base in a rapidly expanded pipe has been investigated. Tiny jets our in number of 1 mm
diameter are positioned at the interval of 90 degrees at 6.5 mm from the central axis of the main jet. The Mach numbers of the abruptly
expanded flows studied for base pressure range from 1.1 to 3 and the obtained wall pressure distribution is depicted for Mach number 1.6
and 1.8 respectively. Axi-symmetric round brass tubes were used to join jets; and cross-sectional area of those tubes are 2.56. L/D ratio of
the broadened pipe was differed from 1 to 10 and NPR was shifted from 3 to 11. Notwithstanding, the outcomes displayed were for Low
L/D values of 4, 3, 2 and 1 individually. Also, when the stream was released to the pipes of the given area ratios, it stayed connected with
the channel divider for all the inertial levels and the NPRs tried in the present case. Further it is understood that level of expansion assumes
a noteworthy part to choose the pressure at the base and its control adequacy. At whatever point, the stream is over expanded, it leads to
the formation of an oblique shock at the nozzle lip, prompting improvement of the pressure in the base locale. Shock waves formation,
reflection and recombination proceeded till the pressure winds up noticeably environmental and seen that the stream stays intact for low
L/D ratio of 4. Very small scale (micro) jets proved to fit in as controllers for the base pressure.

Keywords: Flow; Control; L/D Ratio; NPR; Nozzle Flow; Pressure

working potential, the exhaust nozzle design caters two major func-
1. Introduction tions. In the first place, nozzle section configurations are varied to
achieve control over propulsive system back pressure to attain per-
The Convergent Divergent (CD) nozzle is a noteworthy design ar- fect design conditions. Second, the design converts potential energy
rangement inside a jet propulsion framework that assumes an indis- of the expanding gas to kinetic energy by accelerating the exhaust
pensable part for a vehicle working under supersonic conditions. gas, which accomplished by efficiently expanding the gases to the
The profile of the nozzle is important for designing a nozzle setup ambient pressure (Gamble, Terrell, & DeFrancesco, 2004). S.
for aerospace vehicles flying at higher Mach numbers to limit base Brinkhorst et al (Brinkhorst, von Lavante, & Wendt, 2017) investi-
drag and expand the fumes gas supersonically to augment its poten- gated the choked flow condition in Venturi nozzles with cavities.
tial. Execution of the vehicle predominately relies on upon the setup Shock wave as well as the turbulent boundary layer collaborations
of the divergent area, which guarantees that the heading of the es- happen in a wide range of aerospace devices, for example, super-
caped gases is in reverse direction, as any sideways parts would not sonic wind tunnel diffusers, supersonic inlet diffusers, inlet/com-
add to thrust (Ekanayake, 2013). The gas expansion through a CD bustor isolators, supersonic ejectors, and shock tubes (Hyde,
nozzle from subsonic to supersonic conditions, the flow encounters Escher, & Roddy, 2003; Ikaza, 2000; Sutton & Biblarz, 2016). It is
many distinct physical interactions which increases kinetic energy generally identified with the system efficiency or performance in
including flow separation, flow unsteadiness, flow blending, Shock terms of total pressure loss, flow instability, and other uniquely re-
Induced Boundary Layer separation and Mach shock Diamonds (M quired controls. Many flow devices frequently require a proper con-
Buoni, D Dietz, K Aslam, & VV Subramaniam, 2001; M Buoni, D trol of the Shock wave-boundary layer communication to fill the
Dietz, K Aslam, & V Subramaniam, 2001; Dutton, Herrin, Molezzi, outline need. The stream is decelerated through a progression of
Mathur, & Smith, 1995). Some of these interactions may prompt to shocks. All the interactions linking the boundary layer and the
pressure loss; accordingly lessen the entire thrust created by the shock wave that builds the stagnation pressure losses. Wu and New
nozzle. Moreover, when nozzle flow is under over-expansion or un- (Wu & New, 2017) studied the high speed jets coming out from
der expansion, the thrust loss because of Mach shock Diamonds bevelled nozzles with divergence angles of 3 and 6 degrees at Mach
makes it ineffective (Lipfert & Fruchtman, 1972). Previous exami- 1.5. The matching of Numerical and results are good, where the
nations have demonstrated that flow phenomena happening inside shape of the shock wave was either in the form of triangle and rec-
(shocks and flow separations) and outside (Mach shock Diamonds) tangle. They are not like diamond shock which are normally pro-
to the nozzle still result in performance downsides and are still chal- duced by the circular nozzle. Moreover, the shear layer exiting from
lenging (Mates & Settles, 2005; McLellan, 2006). For achieving its the nozzle will go through the process of deflection due the pressure
relaxation for short and long lip area. The jet exhaust deflects in the
Copyright © 2018 Zakir Ilahi Chaudhary et. al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
 International Journal of Engineering & Technology 237

direction of the large and small nozzle lip regions under the influ- examination to study efficacy of the mechanism assess the effec-
ence of adverse and favourable pressure gradient. tiveness of the tiny jets.
For the improvement of system efficiency, different techniques can The present work is mainly concerned with air jets expanding into
be utilized to control and to alleviate the negative impacts of base ducts with dynamic control with tiny-jets. The nozzle proportion
flow (Fig. 1). ratio (NPR), the Mach number, and L/D ratio were dealt as inde-
pendent parameters.

2. Experimental facility
The analyses were completed utilizing the test set up shown in fig
2 at the Supersonic Aerodynamics Research Laboratory (HSAL),
Mangalore Institute of Technology. Figure explains nozzle exit
houses eight holes out of which four (marked c) were used to blow
and rest for base pressure measurement. Marked holes help to con-
trol base pressure by the process of blowing and utilizing the pres-
sure from pressure regulating tank by engaging a tube interfacing
to the tank and pressure gaps (c). Pressure taps and instrumentation
Fig. 1: Schematic of the Flow Interaction between the Free Stream Flow were same as in case of Ref. [2017].
and the Jet Exhaust around the Aft-End of the Missile/Launcher Tail (Kop-
penwallner, Rammenzweig, & Strunck, 1981).

Several theories focused on improving the base pressure and deflat-

ing negative effects caused due to free stream-exhaust interaction
by altering wake flow field (Viswanath, 1996). The control methods
used were both active and passive. (Tripathi, Manisankar, &
Verma, 2015; Viswanath, 1996). At transonic speeds, the boat-tail
drag offsets the decrease in base drag. (Liang & Fu, 1991) studied
the base drag reduction by passive methods. Later, Hsiung and
Chow found that using a cavity with compartments through splitters
underneath the porous wall increases the drag reduction efficiency
(Hsiung & Chow, 1995). (Ibrahim & Filippone, 2007) studied the Fig. 2: A Schematic of Experimental Set-Up.
base flows for optimum porous media. (Chaudhuri & Hadjadj,
2016) carried out experiments to investigate the shock induced tran-
sient flow through a planar nozzle mimicking a shock tube experi-
mental setup at shock Mach number Ms = 1.86. It is observed that
the initial flow-field greatly affected by the separation point and the
resulting oblique shock structure. Next big task is to decrease base
pressure fluctuations using the above discussed control methods.
These fluctuations are related with turbulent separated flow often
and also to alleviate the dynamic loading of missile’s nozzle struc-
ture (Deprés, Reijasse, & Dussauge, 2004).
The recent design of aerospace vehicles will be greatly influenced
by the existing dynamic flow control at our disposal for jet inlet and
exhaust systems, lifting surfaces including high-lift devices, thrust
monitoring, cavity flow/acoustics, impingement jet noise reduction,
and propulsion devices such as jet engines and rockets (El-Alti, Fig. 3: A View of Suddenly Expanded Duct with Pressure Tapings.
2012; Jahanmiri, 2010; King, 2010; Washburn, Gorton, & Anders,
2002). 3. Results and discussion
The experimental investigation to regulate the flow in the separated
area by tiny jets by Khan and Rathakrishnan (Sher Khan & Ratha- The result analysis procedure was followed as in case Ref. [2017].
krishnan, 2002; S. A. Khan & Rathakrishnan, 2004). Mousavi and The dimensionless base pressure as a function of inertia level at
Roohi (Mousavi & Roohi, 2014) carried out 3-D CFD analysis to various Nozzle Pressure Ratio’s and L/D 4, 3, 2 and 1 respectively
assess the formation of shock train in case of compressible flow are presented in figures 4 ((a) to (d)).
passed a C-D nozzle to evaluate the number of shocks, the location,
and their behavior. From their study they found that RSM turbu-
lence model is capable of predicting the authentic location of the 4. Base pressure distribution
shock waves and they agree well with the experimental results. The
results also indicated the trend of the shock train in the C-D nozzle Comparative pattern followed for lower duct length as 4 and 3, as
from the numerical simulation and Weiss et al. experiment. observed from Figs. 4 (a) and 4 (d). The essential distinction among
McNally et al (McNally et al., 2015) investigated the active flow L/D less than 4, the base weight expects extensively higher values
control on a flat-back ground vehicle model, for aerodynamic drag at M = 3. This increase of base pressure turns out to be substantial
reduction by employing small scale, steady jets (micro jets). and tends to wind up noticeably nearly does not dependent of NPR
Sundararaj et al (Sundararaj, Yuvaraj, Gopinath, & Elangovan, as L/D reduces to low values. This is as a matter of concern when
2014) investigated the Passive control methods like protruding ribs the abrupt duct is beneath some restricting value which is indeed
in the control of suddenly expanding flows. From the computational needed for the stream to re-join and flow downstream is not availa-
analysis for the suddenly expanded axisymmetric flows, passive ble. Due to this phenomena it may lead in non-development of the
controls in the form of annular ribs have been found to reduce the base vortex which generally will be situated at the base.
base pressure significantly, compared to without control. Khan et Fig. 4 (a) presents results for L/D = 4. It is seen that as long as the
al. (Bashir, Baig, Ismail, & Khan; Chaudhary, Shinde, Bashir, & flow remains over expanded the base pressure continue to decrease.
Khan, 2017; S Khan, Bashir, & Asadullah, 2016) did experimental
 238 International Journal of Engineering & Technology

Once the flow has become either correctly expanded or under ex- length is the defining factor for base pressure when flow recircu-
panded the base pressure begins to increase and control is most ef- lates. The minimum value of this length for sonic and subsonic re-
ficient. Similar results are shown in Fig. 4 (b) for L/D = 3. As for gimes as proven by Rathakrishnan and Sreekanth (Rathakrishnan &
L/D =2 is concerned, trend sights similarities as of higher L/D’s Sreekanth, 1995) is L/D = [3]
with minimal control effectiveness. Fig 4(d) which is in contrast
from results of higher L/Ds.
5. Wall pressure distribution
By simulating static pressure across the enlarged duct walls of base
flows, a huge challenge came in form of the periodic nature of pres-
sure field. This problem also comes under scanner in this research
as study embarks on to investigate how effectively active control
techniques can impact wall pressure field. Experimental investiga-
tions were done with and without controls to understand this wall
pressure distribution and distributions were presented in figures 5
(a) to 6 (h). Analyzing these trends of the outputs, it is conclusive
that no significant changes were detected for both with control and
without control. Thus implying, active control barely influences
wall pressure field distributions to vibrate vigorously. This is a pre-
cious advantage as oscillations are one of the major concerns en-
countered in using control for base pressure fields.
Fig. 4: A).
5.1. Mach number: 1.6

For Mach number 1.6, Figure 5 (a) to Figure 5 (h) depict the wall
pressure distribution as a function of X/D, flow field is showing
oscillatory nature for NPR = 9 and 11. These NPRs come under the
limit of under-expansion.
For Mach number 1.8, correct expansion occurs at NPR = 6.4. For
low NPRs [3] and [5] we find that the graphs are not showing any
oscillatory nature. This may be since these NPRs lie outside the
limit of under-expansion. In all graphs, wall static pressure is reach-
ing close to atmospheric pressure at the exit of the enlarged duct.
For L/D = [3], at NPR = [7] and [9] micro-jets are effecting the flow
field but are not aggravating the oscillations, which is a major ad-
vantage of this control. For NPR = [5], wall static pressure reaches
atmospheric pressure (which is also the back pressure) very rapidly
and remain close to atmospheric for remaining (about 50%) length
of the duct.
Fig. 4: B).
The case L/D = [4] and NPR = [9] for which maximum base pres-
sure gain is achieved shows no significant effect on the wall pres-
sure field. It is seen that control is reducing the oscillations for this
case towards the exit of the duct.
For NPR = [9] and L/D = [2] and three we find that the re-attach-
ment length has increased by applying controls. This is what we are
exactly looking for. Increase of re-attachment length is favorable
because this will decrease the strength of vortices at the base region.
Hence, we can obtain higher pressure at the base. For NPR = 11 and
L/D = [2] we again find that re-attachment length has increased by
applying controls. For M = 1.87 correct expansion occurs at NPR =
6.4. So, when we have low levels of under-expansion, i.e., at NPR
= 9 and 11, for L/D = two then again, we obtain re-attachment point
at about 70% of the length of the enlarged duct.
Fig. 4: C).

Fig. 4: D).
Fig 4: Base Pressure Distribution at Different Nozzle Pressure Ratio is with
Fig. 5: A).
and without Micro Jet.

Length to diameter (L/D) proves to be the pivot in achieving control

over base pressure with the help of micro jets. Also, reattachment
International Journal of Engineering & Technology 239

Fig. 5: B). Fig. 5: F).

Fig. 5: C).

Fig. 5: G).

Fig. 5: D).

Fig. 5: H).
Fig. 5: Wall Pressure Distribution at Mach number with and with Microjet
Control.

5.2. Mach number: 1.8

For M = 1.8, the wall pressure distribution is shown from Figure 6

(a) to 6 (h) respectively. From the observation, it should be noted
that the control has got no adverse effect on the static wall pressure
flow field. With this it can be taken
that the micro-jets can serve as base pressure controller without im-
posing any adverse effect in the pressure field in the enlarged duct.
Fig. 5: E). Like above mentioned discussion, to quantify the effect of control
on wall pressure distribution Pw/Pa for the two cases, namely in the
presence and absence flow regulation have been observed. It is ob-
served from the results that, for L/D =4, the control either does not
affect the static pressure (as in the case of Figs. 6 (a) and 6 (b). For
L/D = 3, at NPR 7 it is seen that, there is some considerable influ-
ence of the control taking the wall pressure slightly lower than that
for without control in the vicinity of the base region extending up
to X/D = 4.
 240 International Journal of Engineering & Technology

Similarly, for correctly expanded flow and under expanded flow,

this condition is found, the shear layer which is expanding freely
from the nozzle is strongly influenced by the oblique shock stand-
ing at the nozzle lip. Therefore, flow exiting from the nozzle will
tend to deflect in the direction of the shock. Under such circum-
stances, if micro-jets are activated the entrainment of the micro-jets
is bound to carry some mass from the surrounding. This may be the
case for the lower wall pressure experienced when the control is on.

Fig. 6: E).

Fig. 6: A).

Fig. 6: F).

Fig. 6: B).

Fig. 6: G).

Fig. 6: C).

Fig. 6: H).
Fig. 6: Wall Pressure Distribution at Mach number with and without Micro
Jet Control.

For few combination of variables, it is also observed that when the

micro jets are activated it results in decrease of base pressure and
hence assumes significantly lower values than those without con-
trol. A peculiar tone was generated making the jet almost silent and
the wall pressures become highly oscillatory for such cases. This is
Fig. 6: D).
in good agreement with the observation of Anderson and Williams
(Anderson & Williams, 1968) who stated that reduction in noise is
associated with the decrement in the pressure at base. The observa-
tion of Anderson and Williams has revisited.
 International Journal of Engineering & Technology 241

(A) shock structure. Paper presented at the 35th AIAA Thermophysics

Conference.
[5] Chaudhary, Z. I., Shinde, V. B., Bashir, M., & Khan, S. A. (2017).
Experimental Investigation on the Effectiveness of Active Control
Mechanism on Base Pressure at Low Supersonic Mach Numbers In-
novative Design and Development Practices in Aerospace and Auto-
motive Engineering (pp. 197-209): Springer.
[6] Chaudhuri, A., & Hadjadj, A. (2016). Numerical investigations of
transient nozzle flow separation. Aerospace Science and Technol-
ogy, 53, 10-21.
[7] Deprés, D., Reijasse, P., & Dussauge, J. (2004). Analysis of unstead-
iness in afterbody transonic flows. AIAA Journal, 42(12), 2541-
2550.
[8] Dutton, J., Herrin, J., Molezzi, M., Mathur, T., & Smith, K. (1995).
Recent progress on high-speed separated base flows. Paper presented
at the 33rd Aerospace Sciences Meeting and Exhibit.
[9] Ekanayake, S. (2013). Numerical simulation of a convergent diver-
(B) gent supersonic nozzle flow.
[10] El-Alti, M. (2012). Active Flow Control for Drag Reduction of
Heavy Vehicles: Citeseer.
[11] Gamble E., Terrell, D., & DeFrancesco, R. (2004). Nozzle selection
and design criteria. Paper presented at the 40th
AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit.
[12] Hsiung, J.-L., & Chow, C.-Y. (1995). Computed drag reduction on a
projectile using porous surfaces. Journal of Spacecraft and Rockets,
32(3), 450-455.
[13] Hyde, E., Escher, W., & Roddy, J. (2003). Marquardt's Mach 4.5 su-
percharged ejector ramjet high-performance aircraft engine project.
Paper presented at the 36th AIAA/ASME/SAE/ASEE Joint Propul-
sion Conference and Exhibit.
[14] Ibrahim, A., & Filippone, A. (2007). Effect of a Porosity Strength on
Fig. 7: Frequency Spectrum for Area Ratio = 2.56, A) Without Control at Drag Reduction of a Transonic Projectile. Journal of Aircraft, 44(1),
NPR = [7], M = 1.8, L/D = 10, B) Without Control at NPR = [7], M = 1.8, 310-316.
L/D =10. [15] Ikaza, D. (2000). Thrust vectoring nozzle for modern military air-
craft: DTIC Document.
[16] Jahanmiri, M. (2010). Active flow control: a review: Chalmers Uni-
6. Conclusion versity of Technology.
[17] Khan, S., Bashir, M., & Asadullah, M. (2016). An investigation of
From the results and discussions of the present experimental inves- base flow control by wall pressure analysis in a suddenly expansion
nozzle. Journal of Scientific Research and Development.
tigation the following conclusions can be drawn. [18] Khan, S., & Rathakrishnan, E. (2002). Active control of suddenly
The base pressure is governed by the geometrical parameters viz. expanded flows from overexpanded nozzles. International Journal of
the relief of the passage, L/D ratio, nozzle exit inertia level. The Turbo & Jet-Engines, 19(1), 119-126.
Pe/Pa influences the separated region. In other words, it can be [19] Khan, S. A., & Rathakrishnan, E. (2004). Active Control of Suddenly
stated that the NPR, very strongly influences the base pressure. Expanded Flow from Under Expanded Nozzles. International Jour-
When the micro jets are activated they were found to regulate the nal of Turbo and Jet Engines,(IJT), 21(4), 233-253.
flow in the base region, taking the pressure to considerably higher [20] King, R. (2010). Active Flow Control II: Springer.
values compared to that for without control for area ratio 2.56. [21] Koppenwallner, G., Rammenzweig, D., & Strunck, V. (1981).
Rocket exhaust jet flow and upstream flow field interference. Paper
However, there is certain combination of parameters for which the presented at the Experiments and Flow Field Analysis. Congress on
active control affects negatively. Missile Aerodynamics, Trondheim (Norway).
Base pressure is found to increase with increasing Mach number in [22] Liang, S.-M., & Fu, J.-K. (1991). Passive control method for drag
the supersonic regime. From the present study, we can the best com- reduction for transonic projectiles. Paper presented at the ninth Ap-
bination of flow parameters, geometrical parameters to achieve the plied Aerodynamics Conference.
desired results. [23] Lipfert, F., & Fruchtman, I. (1972). Nonuniform flow field genera-
The duct wall pressure distribution, which usually become oscilla- tion for supersonic compressor stator development: Design and pre-
tory when controls are employed, does not get adversely affected liminary results.
[24] Mates, S. P., & Settles, G. S. (2005). A study of liquid metal atomi-
with micro jets. In general duct L/D = 3 appeared to be the limit for zation using close-coupled nozzles, Part 1: Gas dynamic behavior.
base vortex strength manipulation and L/D less than 3 proved to be Atomization and Sprays, 15(1).
insufficient for the flow to re-attach in most of the cases [25] McLellan, J. W. (2006). The effects of diamond injector angles on
flow structures at various Mach numbers. Texas A&M University.
[26] McNally, J., Fernandez, E., Robertson, G., Kumar, R., Taira, K.,
References Alvi, F., Murayama, K. (2015). Drag reduction on a flat-back ground
vehicle with active flow control. Journal of Wind Engineering and
[1] Bashir, M., Baig, M. A. A., Ismail, A. F., & Khan, S. Control of Sud- Industrial Aerodynamics, 145, 292-303.
denly Expanded Flows from Correctly Expanded and under Ex- [27] Mousavi, S. M., & Roohi, E. (2014). Three-dimensional investiga-
panded Nozzles at Supersonic Mach number for Area Ratio 2.56. In- tion of the shock train structure in a convergent–divergent nozzle.
ternational Organization of Scientific Research-Journal of Mechani- Acta Astronautica, 105(1), 117-127.
cal and Civil Engineering (IOSRJMCE), 76-82. [28] Rathakrishnan, E., & Sreekanth, A. (1995). Rarefied flow through
[2] Brinkhorst, S., von Lavante, E., & Wendt, G. (2017). Experimental sudden enlargements. Fluid Dynamics Research, 16(2-3), 131-145.
and numerical investigation of the cavitation-induced choked flow in [29] Sundararaj, M., Yuvaraj, C., Gopinath, S., & Elangovan, S. (2014).
a herschel venturi-tube. Flow Measurement and Instrumentation, 54, Effect of Rib Geometry on the Base Pressure of Suddenly Expanded
56-67. Flows. Indian Journal of Science and Technology, 7(S5), 61-63.
[3] Buoni, M., Dietz, D., Aslam, K., & Subramaniam, V. (2001). Simu- [30] Sutton, G. P., & Biblarz, O. (2016). Rocket propulsion elements:
lation of compressible gas flow in a micronozzle. Paper presented at John Wiley & Sons.
the Proceedings of the 35th AIAA Thermophysics Conference, [31] Tripathi, A., Manisankar, C., & Verma, S. (2015). Control of base
AIAA. pressure for a boat-tailed axisymmetric afterbody via base geometry
[4] Buoni, M., Dietz, D., Aslam, K., & Subramaniam, V. (2001). Simu- modifications. Aerospace Science and Technology, 45, 284-293.
lation of compressible gas flow in a micronozzle-Effect of walls on
242 International Journal of Engineering & Technology

[32] Viswanath, P. (1996). Flow management techniques for base and af-
terbody drag reduction. Progress in Aerospace Sciences, 32(2-3), 79-
129.
[33] Washburn, A., Gorton, S., & Anders, S. (2002). A Snapshot of Ac-
tive Flow Control Research at NASA Langley. Paper presented at the
first Flow Control Conference.
[34] Wu, J., & New, T. (2017). An investigation on supersonic bevelled
nozzle jets. Aerospace Science and Technology, 63, 278-293.