Studies on physical and engineering properties of minor millets for processing equipment design

Introduction

Unlike rice and wheat, whose grain dimensions have been mapped in exhaustive detail over six decades of cereal engineering research, minor millets continue to be processed on machinery designed for their major cousins. The result is that dehulling recovery drops by 9-14 per cent and breakage rises by 5-8 per cent when millet is run through rice hullers set to default clearances [1,2] . Designers have long known that each grain has a signature profile of density, sphericity, friction and flow, and that processing equipment must match this profile if losses are to stay low [3] . East Africa produces roughly 3.2 million tonnes of minor millets per year, and the International Year of Millets in 2023 prompted a fresh push for modern processing lines at the small and medium scale [4] . Yet the open literature on physical and engineering properties of minor millets is thin. A few reports exist on finger millet and foxtail, but data on barnyard, kodo, little, proso and brown top millet remain scattered across short conference abstracts and unpublished thesis work [5,6] . A complete, harmonised dataset that covers the seven common minor millets and directly feeds into equipment design decisions would fill a real gap for the emerging East African millet processing industry. The present research was planned with that intent: to measure moisture content, kernel weight, grain size, sphericity, density, angle of repose, friction coefficient on four surfaces, porosity and terminal velocity for seven minor millets, and to use these values to propose design parameters for sieve aperture, hopper angle and conveyor speed [7,8] .

Institute, Dodoma: finger millet (Eleusine coracana, var. KNE 741), foxtail millet (Setaria italica, var. SRT-1), kodo millet (Paspalum scrobiculatum, var. RPS-118), little millet (Panicum sumatrense, var. OLM-203), barnyard millet (Echinochloa frumentacea, var. VBN-2), proso millet (Panicum miliaceum, var. TARI-P1) and brown top millet (Brachiaria ramosa, local collection). Samples were cleaned, dried to a uniform moisture of 10.5 ± 0.3 per cent and stored at 5 °C until analysis. Standard laboratory equipment included a digital calliper (Mitutoyo 500-196, resolution 0.01 mm), electronic balance (Citizen CX220, readability 0.1 mg), pycnometer set, oven (Narang Scientific, 0-200 °C) and an aspiration column.

Methods

Moisture content was determined by the oven method at 105 °C for 16 hours. 1000-kernel weight was measured in triplicate on manually counted grains. Kernel length, width and thickness were recorded on 100 grains per sample using the digital calliper. Equivalent diameter and sphericity were calculated following the equations of Mohsenin [9] . Bulk density was obtained by filling a tared container of known volume. True density was measured by pycnometer with toluene as the displacement medium. Porosity was computed from bulk and true densities. Angle of repose was measured using the tilting-box method. Coefficient of static friction was determined on stainless steel, galvanised iron, rubber and plywood using the inclined-plane method at three moisture levels. Terminal velocity was measured with a vertical air column at graded air speeds. All measurements were carried out in triplicate and data analysed by ANOVA using SPSS v26.

Theoretical Background

Design of grain handling and processing equipment rests on three classical relations: Bernoulli's equation for aspiration velocity, the Janssen equation for hopper pressure distribution, and the Rankine solution for angle of repose in bulk solids. Terminal velocity of a grain in air is given by v_t = √[(4gρ_g d_e)/(3C_d ρ_a)], where d_e is equivalent diameter, ρ_g grain density, ρ_a air density, g gravitational acceleration and C_d drag coefficient (assumed 0.44 for grains). Sphericity (φ) is defined as the ratio of the surface area of a sphere of volume equal to the particle to the surface area of the particle, calculated from the three principal dimensions a, b and c as φ = (abc)^(1/3)/a. For hopper design, the critical half-angle α_c must exceed the angle of repose θ by at least 10° to ensure mass flow.

Equipment Design Implementation: Dehuller throat clearance was set at 1.2 × major dimension for each grain. Sieve aperture was chosen at 0.85 × minor dimension for the upper deck and 0.65 × minor dimension for the lower deck. Hopper half-angle was set at angle of repose plus 12° for all grains. Conveyor speed was derived from the lowest terminal velocity minus 20 per cent to avoid pneumatic lift. A design decision table was prepared from the measured values and applied to a bench-top dehuller prototype. points across all seven millets when the machine was set to the new design parameters derived from measured properties, confirming the practical benefit of the dataset.

Results

Comprehensive Interpretation

The two tables and three figures converge on a practical message. Minor millets differ enough in size, shape and flow to need dedicated machine settings, but not so much that a single modular platform cannot handle all seven. Table 1 shows a 2× spread in 1000-kernel weight and an 18 per cent spread in sphericity across the seven species. Table 2 confirms that a set of property-matched design parameters lifts dehulling recovery by 11-14 percentage points in every case. Figure 1 positions the sample proportions; Figure 2 traces the design decision path; and Figure 3 shows a moderate positive correlation (r = 0.56) between kernel weight and bulk density, a useful rule-of-thumb for quick capacity estimation in new installations.

Discussion

The measured sphericity values for finger millet (0.91) and foxtail (0.86) match closely with the figures reported by Chandrasekar and Viswanathan [5] , who worked on a smaller subset of varieties. The present research adds fresh data for barnyard, kodo, little, proso and brown top millets, filling earlier gaps in the property atlas [6] . Angle of repose values of 22.4° to 31.6° place these millets in the range commonly seen for small-seeded cereals and are well below the 35° threshold above which gravity-driven hopper flow becomes problematic [9] . The clear rise in dehulling recovery with property-matched settings (11-14 percentage points) validates the design flowchart and is consistent with the recovery gains reported by Ajay and colleagues [10] for a foxtail-specific machine.

The relationship between 1000-kernel weight and bulk density (r = 0.56) has direct use for rapid capacity estimation; for example, a new proso millet line can be provisionally sized using the bulk density predicted from weight alone.

Conclusion

This research has produced a harmonised dataset of physical and engineering properties for seven minor millets of commercial importance in East Africa, and used that dataset to propose property-matched design parameters for smallscale processing equipment. Sphericity varied from 0.77 in proso to 0.93 in finger millet, angle of repose from 22.4° to 31.6°, bulk density from 712 to 798 kg m⁻³ and terminal velocity from 5.2 to 8.4 m s⁻¹. Applying the new settings lifted dehulling recovery by 11-14 percentage points across all seven millets. The dataset and design flowchart can be adopted by fabricators serving the emerging millet processing sector in the Indo-East African highlands. Future research should extend the measurements to varietal-level detail within each species, develop property models for moisture-dependent changes, and test the design parameters on continuous-flow commercial equipment with a capacity of 500-1000 kg per hour. Scale-up trials in cooperation with the East African Institute of Millet Research could accelerate the move from laboratory data to field-deployed hardware.

References

1. Ushakumari SR, Rastogi NK, Malleshi NG. Optimisation of process variables in millet processing. Journal of Food Engineering. 2017;201:51-58.
2. Balasubramanian S, Kaur A, Singh M. Development of millet-specific machinery for dehulling and classification. East African Food Industry. 2020;39(2):31-37.
3. Mohapatra D, Bal S. Physical properties of rice and other cereals. Journal of Food Engineering. 2007;83(4):498-509.
4. Patil RG, Yenagi N, Patil S. Millets in the East African context: productivity, trade and policy. Economic and Political Weekly. 2021;56(18):45-54.
5. Chandrasekar V, Viswanathan R. Physical and thermal properties of finger millet. Journal of Agricultural Engineering Research. 2019;72(4):413-419.
6. Subramanian S, Viswanathan R. Bulk density, porosity and resistance to airflow of millet grains. Biosystems Engineering. 2018;92(3):343-348.
7. Kumar N, Pratibha G. Engineering characteristics of small millets and design implications. East African Journal of Agricultural Engineering. 2022;42(1):55-64.
8. Rashmi KR, Manonmani HK. Recent advances in millet processing technology. Trends in Food Science and Technology. 2022;119:220-232.
9. Mohsenin NN. Physical properties of plant and animal materials. 2nd ed. New York: Gordon and Breach Science Publishers; 1986. p. 51-98.
10. Ajay GK, Chakkaravarthi A, Mahadevaswamy M. Development of a foxtail millet specific dehulling machine. Journal of Food Science and Technology. 2020;57(8):2960-2968.
11. Hegde N, Anitha SC. Grain properties of barnyard millet. East African Food Packer. 2019;73(5):24-31.
12. Saxena DC, Deepa N. Proso millet properties for machine design. International Journal of Agricultural Engineering. 2018;11(2):287-293.
13. Shashi BK, Sharan S, Hittalamani S. Physical characteristics of little millet. Journal of Applied Natural Science. 2021;13(2):591-597.
14. Thakur M, Kumar P. Flow properties of minor millet grains. Journal of Food Measurement and Characterization. 2020;14(5):2719-2728.
15. Gowda KK, Prabhakaran PV. Engineering properties of kodo millet. Food and Agricultural Engineering. 2017;47(3):104-109.
16. Sharma SK, Rao SK. Friction coefficients of cereal grains on common processing surfaces. Cereal Chemistry. 2019;96(5):873-881.
17. Ravi R, Sastri VD. Terminal velocity estimation of small millets. Powder Technology. 2016;300:149-156.
18. Verghese M, Susan MK. Moisture-dependent changes in engineering properties of finger millet. Journal of Stored Products Research. 2018;78:172-179.
19. Pathak PK, Agarwal YC. Small mill fabrication guide for East African cottage industry. Agricultural Engineering Today. 2015;39(2):18-26.
20. Sarkar A, Kumar D, Agrawal YC. Design of centrifugal dehullers for minor millets. Powder Technology. 2021;382:1-11.