MEASUREMENT OF DROPLET SIZE DISTRIBUTION CHARACTERISTICS OF A SPRAY DRYER-ROTARY ATOMIZER USING PHASE DOPPLER ANEMOMETRY TECHNIQUE

Gholam Reza Chegini¹, Behzad Bashiri², Mohamad Ashjaei^2
1 Department of Mechanical Engineering of Biosystems, College of Aboureihan, University of Tehran, Tehran, Iran
² Department of Agrtechnology, College of Aboureihan, University of Tehran, Iran

ABSTRACT

Experimental tests were conducted for measuring spray size distribution characteristic and power consumption of a spray dryer rotary atomizer using Phase Doppler Anemometry (PDA). For different operating condition of rotary atomizer: 7000–26000 rpm disk speed and 20–80 l/h feed flow rate, the main parameters of spray such as D_v0.1, D_v0.9, SMD and VDM were measured. Statistical analysis indicated that feed flow rate and atomizer speed, except RSF value, are significant effect on spray characteristics. At constant atomizer speed, by increasing feed flow rate from 20 to 80 l/h, SMD and D_v0.1 decreased, VMD unchanged, and D_v0.9 increased. Also, at constant feed flow rate, by increasing atomizer speed, all spray characteristic of D_v0.1, D_v0.9, VMD and SMD decreased. With increasing the atomizer speed and feed flow-rate, obviously power consumption increased.

Key words: spray drying, power consumption, PDA, spray characteristic.

INTRODUCTION

In spray dryers a liquid or suspension is atomized by an atomizer and the droplets are dried out by hot gas. Quality of dried powder is affected by liquid specifications and operating spray dryer variables. One the most important operating variables are atomizer characteristics. There is an extensive range of atomizers utilized for spray formation in industry, and the advantages and disadvantages of each one have been well documented [7]. The rotary atomizer provides versatility in that the droplet sizes of liquid can be easily controlled by simply adjusting the atomizer rotational speed. It is unique in that it permits the simulation of both droplet linear velocities and shearing forces [5]. Rotary atomization processes in spray dryer have extensively been studied [3,9,10,11].

One obstacle to these studies is intrinsic properties of the atomizer. Though the fundamental properties of rotary atomizers and their utilization in both industrial and research applications have been studied, each atomizer has its own inherent characteristics which stem from the form of rotating body and its dimensions [7]. There are many droplet size analyzers that most of them use optical methods for measurement of spray characteristic. Optical analyzers are typically non-intrusive and do not influence the spray behavior during testing. PDPA (phase Doppler/Particle Anemometry) or PDA are flux sampling instruments. In a study, an on-line PDA was employed to observe and monitor the atomization result based on the characteristic particle properties [6,18]. Also, spray measurements were obtained using the PDA techniques for the characterization of swirl atomizers [14]. In 2006, a novel way to use PDA for characterization spray from rotary wheel atomizer were described [17].

The Other important factor for evaluating rotary atomizers in economical aspect is needed power to atomize feedstock. Since rotary atomizers have a great industrial application there weren't enough data about power consumption along with their works. By determining the exact value of consumed power for each feedstock in a rotary atomizer it can be possible to figure out the overall costs more exactly. In general, empirical equations show that droplet size is influenced by atomizer speed, physical properties of liquid, feed flow rate, and intrinsic properties rotary body [2,4,12,15]. Therefore, to begin demonstrating droplets distribution characteristic in rotary atomizer, this parameters have been studied under various operating conditions. In this study droplets size characteristics and consumed power of rotary atomizer with pin-wheel for water is studied.

MATERIALS AND METHODS

Experimental test were conducted by rotary atomizer test apparatus that is shown in Figure 1. It is includes: power analyzer, rotary atomizer, analyzer of PDA, signal processor and computer.

Rotary Atomizer. otary atomizer contain a wheel that mounted on a motor (710 watt model Dewalt rotor, 120 V, 11 amps, 27,000 rpm), a feeding system and atomizer drive case. Wheel has 50 mm diameter and 8 mm outlet thickness and has 11 semi conical pins were located peripherally. Figure 2 shows the design of wheel.

In rotary atomizer: wheel speed was controlled by a speed controller and measured by a tachometer (Prova RM-1501). Flow was peripherally flowed to center of wheel through the liquid distributer, and was measured by a flow-meter (range of 10 to 100 l/h). Power Analyzer (Lutron 801) was used for measuring of consumption power of rotary atomizer. Water was used for all experiments.

Phase Doppler/particle Anemometry (PDA). The PDA analyzer consists of a transmitter, receiver, signal processor and computer. The PDA uses a low-power laser that is split into two beams or four beams for a 2-dimensional system. By using a beam splitter and frequency module; the laser beams intersect again at a point referred to as probe volume. When a drop passes through probe volume, scattered light forms an interference fringe pattern. The scattered interference sweeps past the receiver unit at Doppler difference frequency, which is proportional to drop velocity. The spatial frequency of the fringe pattern is inversely proportional to drop diameter. A data analysis routine is used to convert the raw drop count into a meaningful drop size distribution. The PDA able to measure droplet sizes from 0.5 to 10,000 µm using various optical configurations, that is ideal for complete spray evaluation. The main parameters for study of size distribution are droplet diameter D_v0.1, D_v0.9, D₃₂ and D_v0.5. they are the best items for evaluation of individual drops drift potential [1, 8]. Table 1 is shown method for identification of these parameters.

METHODS

A slit in trough of test stand was constructed to prevent any obstruction of the Dantec's laser diode [16]. The analyzer's laser beams intersection was located 150 mm from wheel edge. For each experiment the analyzer was configured with following options: 600 mm focal length lens, 10 mm laser beam diameter, long bench option, and continuous mode operation. When wheel reached a steady rotational speed, liquid was transfer to center of disk at a constant feed flow rate approximately 30–45 s. Each droplet system was studied at six rotational speeds (7000 to 27000 rpm) and four different feed flow rates (20, 40, 60 and 80 l/h). Parameters of power consumption, and droplet size distribution characteristic (D_v0.1, D_v0.9, D₃₂ and VMD, RSF) for each test were measured and calculated. Experiments were conducted three replications.

RESULTS

The droplet size distribution results and consumed power along with the atomizer operating parameters have been compared. Table 2 shows the effect of operating variables of rotary atomizer: feed flow-rate and speed on droplet distribution characteristics and power consumption. Results indicate that feed flow rate and speed are significant effect on in level of parameters, except RSF value that speed was not significant.

Table 3 shows the effect of different feed flow rate on spray characteristic and consumption power. As can be seen in Table 3, for all spray characteristics at low feed flow-rate (20 and 40 l/h) there are not significant differences between them, but for higher feed flow rates, spray characteristics have significant difference together.

Table 4 shows the effect of different atomizer speed on spray characteristic and consumption power. Results indicate for all spray characteristics, except RSF, there is significant difference in 1% level. Also, there isn't significant difference between mean of speed upper than 12000 rpm for spray characteristic. The effect of atomizer speed and feed flow rate on droplet size distribution characteristics is shown in Figure 3.

Results indicated at a constant atomizer speed, by increasing feed flow rate from 20 l/h to 80 l/h, SMD and D_v0.1 decrease, VMD remains essentially unchanged, and D_v0.9 increases. Previous reports indicate slight increases in droplet size due to increasing feed flow rate [4, 5, 15] Also, at a constant feed flow rate, by increasing atomizer speed, all spray characteristic of Dv_0.1, D_v0.9, VMD and SMD decrease. Decreasing of SMD resulted that spray has been fineness. Changes in feed flow rate have been much less impact on droplet size distribution than atomizer speed [4, 13].

The amount of VMD in most levels is smaller than 100 µm. According to ASAE S327.2 DEC95 standard, the produced spray by atomizer is known as mist. Figure 5e, indicates that by increasing of atomizer speed, RSF decreased, therefore can results that spray distribution have been uniform. Droplet size distribution is typically expressed by the size vs. the cumulative volume percent. Droplet size distribution of sprays from atomizer wheel is the best represented on log-probability figure. Figure 4 shows the effect of atomizer speed on droplet size distribution at various feed flow rates.

The figure that represented the lognormal drop size distribution and fitted Rosin-Rammler curve is acquired by Size-Ware software of PDA apparatus. As disk speed increases from 7000 rpm to 26000 rpm, droplet size distribution becomes more uniform and D_v0.1, SMD, and D_v0.9 decrease respectively from 93.44, 127.34 and 192.24µm to 54, 69.8 and 94.461 µm at feed flow-rate of 20 l/h. The results parallel those found experimentally for other rotary atomizers in ability to generate a range of size distribution, and thus create different aerosol conditions by simply changing speed [4, 12, 13]. The results also show that changes in atomizer speed have greater impact on larger droplets; however this impact is well recognizable in high feed flow rates (Fig. 4b).

Figure 5 shows the effect of atomizer speed on power consumption in different feed flow-rate. With increasing the atomizer speed and feed flow-rate, obviously power consumption increases. As can be seen in figure 4, at low feed flow-rate there are not significant differences between levels, but these differences is very clear in higher speeds.

CONCLUSIONS

Experimental test was conducted for measuring spray size distribution characteristic and power consumption of a rotary atomizer. Statistical analysis indicated that feed flow rate and speed are significant effect on spray characteristics, except RSF value that speed was not significant at a constant atomizer speed, by increasing feed flow rate from 20 l/h to 80 l/h, SMD and D_v0.1 decrease, VMD remains essentially unchanged, and D_v0.9 increases. At a constant feed flow rate, by increasing atomizer speed, all spray characteristic of Dv_0.1, D_v0.9, VMD and SMD decrease. With increasing the atomizer speed and feed flow-rate, obviously power consumption increases.

ACKNOWLEDGMENT

Thanks to Associate Dean for Research Affairs, College of Aboureihan, Mechanical department of Technical campus, University of Tehran and Iranian National Science Foundation (INSF) for financial supports.

REFERENCES

1. Arnold A.C., 1990. A comparative study of drop sizing equipment for agricultural fan-spray atomizers. J. Aerosol Sci. Technol., 12, 431–445.

2. Boize L.M., Dombrowski N., 1976. The atomization characteristics of a spinning disc ultra-low volume applicator. J. Agric. Eng. Res, 21, 87–99.

3. Fraser R.P., Eisenklam P., Dombrowski N., Hasson D., 1963. Drop formation from rapidly moving sheets. AIChE J., 8, 672–680.

4. Glahn A., Busam S., Blair M.F., Allard K.L., Wittig S., 2001. Droplet generation by disintegration of oil films at the rim of a rotating disk. J. Eng. for Gas Turbines and Power. 124, 117–124.

5. Hewitt A.J., Robinson A.G., Sanderson R., Huddleston E., 1994. Comparison of the droplet size spectra produced by rotary atomizers and hydraulic nozzles under simulated aerial application conditions. Environ. Sci. Health, 4, 647–660.

6. Husted B.P., Petersson P., Lund I., Holmstedt G.O., 2009. Comparison of PIV and PDA droplet velocity measurement techniques on two high-pressure water mist nozzles. Fire Safety J., 44, 1030–1045.

7. Lefebvre A. H., 1988. Atomization and sprays. Hemisphere Publishing Corporation. Philadelphia.

8. Liu X.F., Doub W.H., Guo C.N., 2010. Evaluation of droplet velocity and size from nasal spray devices using phase Doppler anemometry (PDA). Int. J. Pharmac., 388, 82–87.

9. Masters K., 1991. Spray drying handbook. Longman Scientific & Technical. UK.

10. Masters K., 2002. Spray drying in practice. Spray Dry Consult International. Denmark.

11. Matsumoto S., Crosby E.J., Belcher D.W., 1985. Rotary atomizers. Performance understanding and prediction. Institute of Energy; Third Int. Conf. on Liquid Atomization and Spray Systems. London, 1–21.

12. Ryley D.J., 1959. Analysis of a polydisperse aqueous spray from a high-speed spinning disk atomizer. J. Ph. Korthals Altes, 10, 180–186.

13. Salyani M., 1988. Effects of flow rate and rotational speed performance of two rotary atomizers. Pesticide Formulations and Application Systems, 18, 70–79.

14. Soltani M.R., Ghorbanian K., Ashjaee M., Morad M.R., 2005. Spray characteristics of a liquid-liquid coaxial swirl atomizer at different mass flux ratios. Aerospace Sci. Technol., 592–604.

15. Sternowsky S., Shulte G., Guardani R., Nascimento C.A.O., 2002. Experimental study and neutral network modeling of the ligament disintegration in rotary atomization. Atomization and Sprays, 12, 107–121.

16. Willauer H.D., Hoover J.B., Ananth R., Hoover J.B., Mushrush G.W., Williams F.W., 2006. Critical evaluation of rotary atomizer. Petrol. Sci. Technol., 24, 1215–1232.

17. Willauer H.D., Hoover J.B., Mushrush G.W., Williams F.W., 2005. The construction of an improved automated atomizer for evaluating jet fuel flammability. J. Petrol. Sci. Technol., 23, 1075–1082.

18. Ziesenis J., Bauckhage K., 2003. Spray forming. Controlling the atomization result with regard to particle properties. Part. Part. Syst. Charact. 20, 290–297.

Accepted for print: 29.09.2010

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.