Home Technology Operation Study of Miniature Air-blast Atomizer under Very Low Liquid Pressures
Article
Licensed
Unlicensed Requires Authentication

Operation Study of Miniature Air-blast Atomizer under Very Low Liquid Pressures

  • Igor Gaissinski EMAIL logo , Yeshayahou Levy , Daniel Kutikov , Valery Sherbaum and Vladimir Rovenski
Published/Copyright: April 12, 2017
Become an author with De Gruyter Brill

Abstract

Miniature air blast atomizers are intended for use in small jet engines. The typical miniature atomizer with rotating air outside the pressure swirl nozzle was investigated by theoretically and experimentally. The equations which describe the flow of rotating hollow liquid jet in a gaseous environment, where there is difference between outer and inner pressures of the liquid film were developed. The effects of surface tension, gravity forces, and inner-outer pressure difference on liquid film modes and its disintegration were considered. A set of the relevant equations was solved numerically. The initial conditions for the equations: pressure distribution inside and outside of the ‘onion’ as function of outer rotating air and liquid parameters, were obtained experimentally. The numerical results were verified by the experimental study. It was shown that exists two shapes of liquid film nearby nozzle exit – ‘onion’ and open film. Shape of the liquid film for the given nozzle pressure drop depends mainly on air pressure difference outside and inside the film. Two resulting spray modes are distinguished in droplet characteristics drastically. The results can be applied for estimation of the operational parameters of air-blast atomizers that operate with very low liquid pressure drop.

Funding statement: The authors are grateful to ISRAEL SCIENCE FOUNDATION (ISF) for its support (ISF Grant No 661/11).

Appendix 1

In the case of εRe1=0,Eu=0,Fr the problem (5) – (7) leads to the following system of differential equations:

(43)dVτds=Vθ2VτRsinψ,dVθds=VθRsinψ,dψds=cosψRVθ2/VτWe1RVτWe1R,dRds=sinψ,dxds=cosψ

with the boundary conditions

(44)x|s=0=0,R|s=0=1,Vτ|s=0=1,Vθ|s=0=VΩ,ψ|s=0=ψ0

Substitution the fourth equation into the first and the second ones gives

VτdVτds=Vθ2RdRds,1VθdVθds=1RdRdsdψds=cosψRVθ2/VτWe1RVτWe1R,dRds=sinψ,dxds=cosψ

Hence

Vτ=12C2/R4+C1,Vθ=C/Rdψds=cosψRVθ2/VτWe1RVτWe1R,dRds=sinψ,dxds=cosψ

Using the boundary conditions, one can obtain the following:

(45)Vτ=12VΩ2(1R4)+1,Vθ=VΩ/Rdψds=cosψR(VΩ2/R2)/[12VΩ2(1R4)+1]We1R12VΩ2(1R4)+1We1R,dRds=sinψ,dxds=cosψ

Proposition 1

In the case when R=const,ψ=const,Vτ=const, and Vθ=const is the solution to the problem (45) – (44) it is necessary and sufficient to have the following boundary values: ψ0=0, VΩ=We1/2.

Proof: 1

Necessity: If R=const,ψ=const, then follow initial conditions (44) R1, ψ=0, and so the system (45) has the following simplified form:

R=const=1,ψ=const=0,Vτ=const=1,Vθ=const=VΩdψds=cosψVΩ2We11We1=0

The result VΩ=We1/2 proves the necessity.

2. Sufficiency: Let VΩ and ψ0 be equal VΩ=We1/2,ψ0=0, and we assume that Rconst, Vτconst, Vθconst, and ψconst is the solution to the problem (45) – (44). Note that the constant functions R=1,ψ=0,Vτ=1, and Vθ=We1/2 also satisfy the problem above as it was proven in the previous point (1). Hence, because of the uniqueness theorem for the Cauchy problem, only the single solution namely R=1,ψ=0,Vτ=1,Vθ=We1/2 satisfies the problem (45) – (44). The last completes the proof.

Appendix 2. Representation of the measured air pressure distribution in the form of an analytic function

Due to the axisymmetric nature of swirl flow, the empirical results are expected to be axisymmetric. Visual inspection of the results revealed a distinctive pattern of a sub-pressure pit on the symmetry axis at some particular distance from the atomizer. Consequently, the following function was selected for further fitting procedure:

(46)F(r,x)=Pmin1[1+exp{a2(rb2)}]11+exp{a1(rb1)}×1[1+exp{a4(xb4)}]11+exp{a3(xb3)}

The suggested function is formed based on an elementary component, named the “logistic function”:

(47)Sx=1+expaxb1

In graphical representation, logistic function has the pattern shown in Figure 32. Consequently, the suggested function F, composed of basic logistic-function elements, attains the following pattern:

  • – The leading coefficient Pmin corresponds to the lowest pressure value within the field – the “pit” pressure;

    – The “pivot point” coefficients bi are evaluated by means of visual examination of the results.

Figure 32: The logistic function.
Figure 32:

The logistic function.

Since bi numerical values are associated with the particular swirler used in the experiments, the “pivot point” coefficients are also defined by means of the swirler diameter, Dmean=Dout+Din/2. This means that the diameter dimension is believed to be the “pivot point” governing factor. In order to define the steepness coefficients ai, a procedure of “least square” approximation was applied:

(48)ijFri,xjPri,xj2=minfa1,a2,a3,a4

where P represents the empirical pressure field.

The resulting analytical function has the form of function F (eq. (46)) with the coefficients placed in Table 8:

To simplify the function and reduce number of coefficients the coefficients ai were fitted separately for each flow rate and then averaged, Table 9. The leading coefficient Pmin, which represents the lowest pressure in the distribution and was presented by the following function

(49)Pminm˙a=33.0m˙a2+6.97m˙a+0.364Pa
Table 9:

Fitted analytical function F, bi coefficients (Dmean=10mm, Figure 8).

b1b2b3b4
0.4Dmean0.4Dmean0.1Dmean1.1Dmean

where m˙a air mass flow rate, g/s.

Finally, the following analytical formula for the pressure distribution, from (46) and Tables 9 and 10, is obtained:

(50)FR,xˉ=Pmin111+expa˜2λR0.41+expa˜1λR+0.4111+expa˜4λxˉ1.11+expa˜3λxˉ+0.1
Table 10:

ai coefficients.

a1a2a3a4
17.9/Dmean28.6/Dmean15.4/Dmean4.4/Dmean

where λ=R0/Dmean,a˜i=aiDmean,R=r/R0,xˉ=x/R0 are our dimensionless variables.

Comparison of measured and analytical approximating function of negative pressure distribution (relatively to ambient) along radius for different air mass flow rates is shown in Figure 33.

Figure 33: Comparison of measured and analytical approximating function of negative pressure distribution (relatively to ambient) along radius for different air mass flow rates downstream distances x1= 2 mm, x2= 5 mm, x3= 8 mm, top-down.
Figure 33:

Comparison of measured and analytical approximating function of negative pressure distribution (relatively to ambient) along radius for different air mass flow rates downstream distances x1= 2 mm, x2= 5 mm, x3= 8 mm, top-down.

References

1. Abo-Serie E, Abdl-Aziz F, Afifi M, Kayed H. Swirl injector nozzle profile for different spatial mass distributions and spray dispersions, ILASS – Europe 2011. 24th European Conf. on Liquid Liquid Atomization and Spray Systems, Estoril, Portugal, Sept. 2011, 1–8.Search in Google Scholar

2. Lefebvre AH. Atomization and sprays. New York: Combustion (Hemisphere Pub. Corp.), 1989:421.10.1201/9781482227857Search in Google Scholar

3. Epikhin VE, Shkadov VY. On the length of annular jets interacting with an ambient medium. Fluid Dyn 1983;18(6):831–838.10.1007/BF01090734Search in Google Scholar

4. Epikhin VE. On the shapes of swirling annular jets of a liquid. Fluid Dyn 1979;14(5):751–755.10.1007/BF01409818Search in Google Scholar

5. Epikhin VE, Shkadov VY. Numerical simulation of the non-uniform breakup of brea capillary jets. Fluid Dyn 1993;28(2):166–170.10.1007/BF01051202Search in Google Scholar

6. Entov VM. On the dynamics of films of viscous and elastoviscous liquids. Arch Mech 1982;34(4):395–407.Search in Google Scholar

7. Yarin AL. Free liquid jets and films: hydrodynamics and rheology. Harlow, NY: Longman Scientific & Technical and Wiley & Sons, 1993.Search in Google Scholar

8. Entov VM, Kestenboim KS, Rozhkov AN, Sharchevich LI. Dynamic equilibrium shape of a film of viscous or viscoelastic liquid. Fluid Dyn 1980;15(2):181–189.10.1007/BF01342605Search in Google Scholar

9. Entov VM, Yarin AL. Dynamical equations for a liquid jet. Fluid Dyn 1980;15(5):644–649.10.1007/BF01089634Search in Google Scholar

10. Nayfeh AH. Perturbation methods. Wiley Classic Library, Wiley–Interscience: USA, 2004:425.Search in Google Scholar

11. Gaissinski I, Rovenski V. Nonlinear models in mechanics: instabilities and turbulence, Lambert: Germany, 2010:700.Search in Google Scholar

12. Kazennov AK, Kallistov YN, Karlikov VP, Sholomovich GI, Investigations into thin annular jets of incompressible liquids, 1970, Nauchn. Tr. Inst. Mekh. Moscow Univ., N 1 (Russian).Search in Google Scholar

13. Chuech SG. Numerical simulations of non-swirling and swirling annular liquid jets. AIAA J 1993;31(6):1022–1027.10.2514/3.11724Search in Google Scholar

14. Kihm KD, Chigier NA. Experimental investigations of annular liquid curtains. J Fluid Eng 1990;112(3):61–66.10.1115/1.2909370Search in Google Scholar

15. Havkin Y. Centrifugal nozzles. Mashinostroenie: Moscow (USSR), 1976:168. (Russian)Search in Google Scholar

16. Rizk NK, Lefebvre AH. Internal flow characteristics of simplex swirl atomizers. AIAA J Propul Power 1985;1(3):193–199.10.2514/3.22780Search in Google Scholar

17. Levy Y, Sherbaum V, Levin D, Ovcharenko V. Air-blast swirl atomizer for small jet engines, GT 2005-68314, ASME Turbo Expo 2005, Power for Land, Sea and Air, June 6–9, 2005, USA.10.1115/GT2005-68314Search in Google Scholar

18. Levy Y, Erenburg V, Sherbaum V, Ocharenko V, Roizman A, The use of methanol as an alternative fuel: droplet formation and evaporation, GT2015-42965, ASME Turbo Expo 2015, June 15–19, 2015, Montréal, Canada.10.1115/GT2015-42965Search in Google Scholar

Received: 2016-11-21
Accepted: 2016-11-22
Published Online: 2017-04-12
Published in Print: 2019-11-18

© 2019 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 19.1.2026 from https://www.degruyterbrill.com/document/doi/10.1515/tjj-2016-0073/html
Scroll to top button