Content uploaded by Ishmael Nartey Amanor
Author content
All content in this area was uploaded by Ishmael Nartey Amanor on Apr 14, 2021
Content may be subject to copyright.
Performance evaluation of mechanical feed mixers
using machine parameters, operational parameters
and feed characteristics in Ashanti and
Brong-Ahafo regions, Ghana
Adusei-Bonsu Micheal
a
, Amanor Ishmael Nartey
b,*
, Obeng George Yaw
c
,
Mensah Ebenezer
a
a
Department of Agricultural and Biosystems Engineering, College of Engineering, Kwame Nkrumah University of Science and
Technology, Kumasi, Ghana
b
Department of Mechanical Engineering, Cape Coast Technical University, Box DL 50, Cape Coast, Ghana
c
Department of Mechanical Engineering, College of Engineering, Kwame Nkrumah University of Science and Technology,
Kumasi, Ghana
Received 24 June 2020; revised 4 March 2021; accepted 23 March 2021
KEYWORDS
Vertical feed mixer;
Horizontal feed mixer;
Coefficient of variation;
Mixing time;
Poultry feed industry;
Ghana
Abstract Mechanical feed mixers are preferred for their higher efficiencies. A wide range of these
feed mixers are available and used in the Ashanti and Brong Ahafo regions of Ghana due to a
higher concentration of poultry farms in these regions. The performances of both local and foreign
poultry feed mixers in the two regions were evaluated. This was carried out as a procedure to iden-
tify the types, measure machine parameters, operational parameters and feed characteristics rele-
vant to the performance of poultry feed mixers. Data collected were analysed using statistical
tools at 95% C.I. From the results, moisture content of the feed samples ranged from 9.21 to
11.08% (% wb), the geometric mean diameter of all samples ranged from 762.74 to 947.27 mm with
the geometric standard deviation range of 2.35–2.50. The two dominating feed mixer types were the
horizontal and vertical. The horizontal feed mixers indicated mixing time of 1.5–10 min, mixing
speed of 40–80 rpm, energy consumption per mix of 0.9–3.8 kWh and blending uniformity with
a coefficient of variation (CV = 7.19%), but relatively high maintenance cost per month. CV less
than 10% is the percent commonly recognized by the feed industry as the cut-off for uniformity of
mix analysis. The vertical feed mixers indicated low maintenance cost, mixing time of 15–30 min,
mixing speed of 250–500 rpm, energy consumption per mix of 0.8–5.5 kWh and CV = 10.30–18.
05%. Given that the horizontal feed mixers were found to have better performance metrics, there
*Corresponding author.
E-mail address: ishmael.amanor@cctu.edu.gh (I.N. Amanor).
Peer review under responsibility of Faculty of Engineering, Alexandria University.
Alexandria Engineering Journal (2021) 60, 4905–4918
HOSTED BY
Alexandria University
Alexandria Engineering Journal
www.elsevier.com/locate/aej
www.sciencedirect.com
https://doi.org/10.1016/j.aej.2021.03.061
1110-0168 Ó2021 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Alexandria University.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
is the need to adapt the design for local process planning and manufacture, while substituting
imported machine parts and components with locally available materials to optimize performance
and minimise maintenance cost for the poultry and feed industries.
Ó2021 THE AUTHORS. Published by Elsevier BV on behalf of Faculty of Engineering, Alexandria
University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).
1. Introduction
Poultry product consumption has seen an increase globally
over the years, particularly chicken. Further, the poultry
industry has also witnessed tremendous improvement over
recent decades to fulfill the increasing demand for economical
and safe consumption of meat and eggs [1]. Within the past
three decades the poultry sector has been growing over 5%
each year compared to 3% for pork and 1.5% for beef and
its stake in world meat production increasing from 15 to
30% [2]. Global demand for poultry products is on the rise,
mostly in developing countries [3]. According to Ravindran
[2], the growth in the demand for poultry product industry is
having an intense effect on the demand for raw materials
and feed. In most African countries including Ghana, various
policies have been advocated to increase the effective utilisa-
tion of mechanical technologies in agricultural production
such as animal feed processing.
The use of mechanised technologies does not come without
a cost compared to the traditional approach [4]. According to
Narrod and Pray [5], technology revolution in the poultry
industry has been very rapid. The change from free-range to
confined poultry operations intensely increased the number
of birds, which could be managed by one farmer. This drift
expedited the substitution of capital for feed and labour in ani-
mal production, and led to technology modification within the
poultry trade, which allowed farmers to extend output per unit
of feed [6]. The use of hand tools like shovel to combine
ground feed ingredient was probably the primary form of
poultry feed mixer. This technique was replaced with manually
operated machines after the industrial revolution in United
Kingdom of Great Britain and Northern Ireland [7].
As mill became multi-storey structures, gravity spouting was
used to direct the meal and grain flow from the highest of the ele-
vator legs. However, gravity spouting could neither reach the
locations desired nor handle the varied textures of grains
selected by mill customers to satisfy their own individual desires.
Horizontal devices were merely shaft with short pegs mounted in
a spiral pattern that moved the grain once the shaft turned,
housed in long wooden boxes. Once electrical motors were
invented, sweeping changes in mill design occurred. Specialised
kinds of conveyors were developed, resembling the feeder. The
development was virtually an equivalent to the material han-
dling conveyor, except that it had been shorter and driven by
a motor with a speed reduction device called a gear box. This
allowed for the achievement of variable speeds that gave mill
operators supplementary flexibility of operation [8].
Feed production for poultry, aquatic life or livestock
involves a series of activities, such as milling, mixing, pelleting
and drying operations [9]. New [10] gave the different kinds of
machinery necessary for the production of different types of
feeds and they include grinders, mixers, conveyors and eleva-
tors, cookers, extruders, driers, steam boilers and fat sprayers.
According to Balami et al. [9], the mixing operation in particu-
lar, is of great significance, since it is the means through which
two or more ingredients that form the feed are interspersed in
space with each other for the main aim of achieving a homoge-
nous mixture that is able to meet the nutritional needs of the tar-
geted poultry, livestock or aquatic life being raised.
Most of the setbacks in feed mixing are due to variations
among feed ingredients in particle shape, size and density.
Feed ingredients with uniform densities and sizes tend to mix
quickly and simply (e.g. cracked or ground grains have densi-
ties the same as that of the seed meals) [11]. Hence, there’s
sometimes little or no issue in getting a uniform mix of those
feed ingredients. Densities of minerals on the other hand are
greater than that of oilseed meals and grains. Variation in
physical form and density of individual feed ingredients makes
the preparation of uniform feed mixtures tough [12].
Mainly, feed mixing can be done either mechanically or man-
ually. The traditional way of using shovel to intersperse feed
constituents into each other on open concrete floor is the main
manual method of mixing feed. The manual method of feed mix-
ing is generally considered ineffective, less efficient and labour
intensive. The mechanical method of feed mixing is attained
by using mechanical feed mixers developed over the years to
overcome the constraints associated with the manual method
[9]. There are a lot of mechanical mixers available for mixing
feed constituents, the selection of which depends mainly on type
of ingredient to be mixed and the efficiency of blending needed.
Brennan et al.[13] noted that irrespective of the type of feed
mixer, the fundamental intention of using a mixer is to achieve a
uniform distribution of the components by means of flow, which
is generated by mechanical means. A sizeable number of
mechanical feed mixers are available in the Ashanti and Brong
Ahafo regions of Ghana due to a higher concentration of poul-
try farms. There are different makes of feed mixers in use across
the two regions, which require a study to evaluate performance
by analysing the feed mixing machine parameters, operational
parameters and feed characteristics. Evaluating the perfor-
mance of the various feed mixers will contribute to optimise
design and manufacture, performance and feed characteristics
for the poultry and feed industries in Ghana and Africa.
2. Materials and methods
2.1. Study area
The study was carried out in the Ashanti and Brong Ahafo
Regions of Ghana (Fig. 1). Brong Ahafo and Ashanti regions
in Ghana are famous for agriculture and agribusiness indus-
tries including poultry farming and feed industry.
Spatial maps of poultry farms were generated using ArcGIS
version 10.2 for both regions from data secured from Ghana
Poultry Farmers Association in the early stages of the research
to have a fair idea of farms distribution in the regions and aid
4906 M. Adusei-Bonsu et al.
in selecting a suitable study area. Much concentration was on
municipalities and districts surrounding the Kumasi metropolis
and Dormaa municipality, because these are where most of the
large scale farms (over 10, 000 birds) are located.
2.2. Study approach
A pre-test study was first conducted by randomly selecting
10 on-farm and commercial mixers in Kumasi and its envi-
rons, and 10 on-farm and commercial mixers at Dormaa
Ahenkro and its environs. For the main work, 50 farms
were captured for the work (24 in the Ashanti Region and
26 in the Brong Ahafo Region), constituting 60% and
81.25% of farms with bird capacity of 35,000 and above,
in the Ashanti and Brong Ahafo Regions, respectively. Feed
samples were collected from 20 feed mixers each, both on-
farm and commercial producers from the Ashanti and Brong
Ahafo Regions.
2.3. Moisture content
The initial moisture content (MC) of the samples on wet
basis were determined by taking three samples from each
mill using the oven drying method. Each sample was dried
at 40 °C until a constant weight was achieved and recorded.
The moisture content of the samples was calculated using
Eq. (1).
W¼wiwf
wi
100 ð1Þ
where,
W = moisture content (%)
w
i
= initial weight of sample (g)
w
f
= final weight of sample (g)
source: (Agyei, 2017)
2.4. Particle size analysis
Two hundred grams (200 g) sample of feed from each replicate
was weighed and oven dried at 40 °C for 6 h to attain uniform
MC [14]. A 100 g sample was measured from the dried one for
particle size analysis to determine ingredient particle size uni-
formity. The experiment was done using Tyler sieves with a
sieve shaker (Retsch GmbH and Co. KG, Germany). The sizes
of screen openings 8, 4, 2, 1, 0.5, 0.25 and 0.125 mm were used.
The sieves were first weighed and their initial masses recorded.
The sieves were stacked on a pan with smaller screen size at the
base and the bigger size up, before placing the stacked screens
on the shaker. A 100 g feed sample prepared was poured into
the top sieve on the shaker, was then covered firmly and the
shaker switched on with the appropriate time set (5 min) at a
shaking frequency of 45 Hz. The weight of particles collected
on each screen including the pan were measured and recorded
by weighing each sieve with its content and the initial weight of
the sieve subtracted from it [15]. The measured weight of par-
ticles collected on each screen were entered onto a spreadsheet
to determine the diameter of the i
th
sieve in the stack, geomet-
ric mean diameter (GMD), geometric standard deviation
(GSD), fineness modulus, and uniformity index of the particles
were calculated using Eqs. (2)–(8)
di¼dud0
ðÞ
0:5ð2Þ
Dgw ¼log1PðWilogiÞ
PWi
ð3Þ
Sgw ¼log1PWilogdilogDgw
2
PWi
ð4Þ
Source: ASAE [16]
Where,
di= diameter opening of i
th
sieve in the stack (mm)
Fig. 1 Map of Ghana showing the study area (generated with ArcGIS version 10.2).
Performance evaluation of mechanical feed mixers 4907
du= nominal sieve aperture size in the next, larger than i
th
sieve (mm)
d0= diameter opening through which particles will not
pass (i
th
sieve), (mm)
Dgw = geometric mean diameter or median size of particles
by size (mm)
Sgw = geometric standard deviation of particle diameter by
mass
Wi= mass on i
th
sieve (g)
Weightretained gðÞ¼S2S1ð5Þ
S
1
= Initial sieve weight (g)
S
2
= Weight of feed retained on sieve (g)
WeightedRetained AðÞ¼weightRetained fð6Þ
f = factor (0, 1, 2, 3, 4, 5, 6 and 7)
Finenessmodulus FMðÞ¼¼
PA
100 ð7Þ
Uniformityindex UI
ðÞ
¼C
10 :
M
10 :
F
10 ð8Þ
C = coarse particles (8–2 mm)
M = medium particles (<2 mm – 0.5 mm)
F = fine particles (0.25 mm)
2.5. Mixer performance testing
2.5.1. Sampling and sample preparation
The sampling was done carefully using a method described by
Stark and Saensukjaroenphon [17]. A 10 g representative sam-
ple was taken from 10 different locations within the mixer (i.e.
4 samples along each side of the mixer and one from each end)
for horizontal mixers whilst in the case of vertical mixers, the
10 samples were obtained during discharge at an interval of
15 s. The samples were ground with a coffee grinder to achieve
uniform particle size.
2.5.2. Uniformity testing
Salt at 0.3% of the totalingredient weight (in kg) was addedto the
feed ingredient during machine mixing. The particle size of the
salt was less than 400 mm as directed by Stark and Saensukjaroen-
phon [17]; this was the standard for conducting a mixer unifor-
mity test using the Quantab Chloride Titrator method (Fig. 2).
The ground samples were tested to determine the mixing
machines efficiency, in terms of their ability to blend the vari-
ous feed ingredients used well per their operation conditions.
The steps used during the experiment were as follows:
I. A 10 g of ground feed sample was weighed into a cup,
90 g of hot distilled water (60 °C) was added to the sam-
ple in the cup. Using a 0.1 g readability scale for both
sample and water.
II. Mixture was stirred for 30 s, allowed to rest for 60 s and
stirred for
III. another 30 s.
IV. Folded filter paper was placed into the cup with solution
and a Quantab strip range of 30–600 mg/L (Hach Com-
pany, Loveland, CO) was inserted into the liquid at the
bottom of the filter paper as shown in Fig. 3.3. The same
lot of Quantab strips were used for all ten samples.
V. The Quantab number at the top of the white peak was read
after the colour of the top band of the strip has changed
from yellow to black, and the Quantab strip reading con-
verted to %NaCl using the chart on the bottle
VI. The %NaCl in the sample was calculated by multiplying
the %NaCl from the table on the bottle (from Step IV)
by 10.
VII. Coefficient of variation (CV) was used to determine mix-
ing uniformity by computing from the results of 10 sam-
ples within a batch. The CV for each batch was
calculated by dividing the standard deviation by the
average value multiplied by 100.
2.6. Data analysis
Data collected from the field were analyzed using the basic sta-
tistical tool pack in Statistical Package for Social Scientist,
while the data gathered from laboratory experiment were anal-
ysed using one-way analysis of variance (ANOVA) at 95%
confidence level, by adopting Fisher’s comparison procedure
with two-side interval, in minitab version 17.
3. Results and discussion
3.1. Types of poultry feed mixers
The mixers are in various types and the ones commonly used
by the farmers are shown in Fig. 4 with their respective
Fig. 2 Quantab strip (Source: Picture by author). Fig. 3 Quantab strip inserted in solution.
4908 M. Adusei-Bonsu et al.
percentage of dominance. Vertical mixers were the most com-
monly used types and all were built locally. Vertical mixers
have a 90% dominance whilst horizontal mixers (paddle and
ribbon) constitute 8% of the total number of mixers sampled.
The ‘Other’ in Fig. 4, represents machine designs that were not
either vertical or horizontal (e.g. slanted, etc.).
Five of the mixers were foreign and 46 were locally manu-
factured ones from the Department of Agricultural and
Biosystems Engineering (DABE), VA Engineering, Supermach
Enterprise (SM) and Science Import and Substitution (SIS)
Engineering.
3.2. Machine makes in the study area
The mixer makes identified during the study are as follows in
Figs. 5–10. Their features and technical differences are in
Tables 1–4 below.
The various mixer performance parameters and specifica-
tions of the four major local manufacturers (all vertical mixers)
and foreign makes (all being horizontal mixers) identified dur-
ing the study are listed in Table 1. From Table 1, the results of
the horizontal feed mixers showed less mixing time and rota-
tional speed. Mixing time is a function of mixer design and
the rotational speed of the auger [18]. This is because horizon-
tal feed mixers have wider and more mixing zones; hence mix-
ing is faster, requiring less time to mix. Horizontal feed mixers
demand high torque to operate efficiently, thus adopting high
horsepower motors (0.9–3.8 kWh energy consumption per
mix) with less speed is the best combination to achieve that
[18]. All the local mixers (vertical types) had almost the same
level/values of performance parameters which conform to
standard design values. Therefore, the difference is about only
the design. Vertical type mixers have only two mixing zones
that are the top and base of the inner tube.
The pie charts in Figs. 11 and 12 show the proportions of
the market that is captured by the various manufacturers of
local feed mixers in the Ashanti and Brong-Ahafo regions.
SIS Engineering dominates the Ashanti Region with a market
share of 54%, whereas VA Engineering also dominates the
Brong Ahafo Region with a share of 88%. Both manufacturers
are more visible in their respective regions. However, Depart-
ment of Agricultural and Biosystems Engineering (DABE)
feed mixers are found in both regions but with little dominance
in both, though it is the second dominant feed mixer in the
Ashanti Region. DABE type of feed mixer is the choice of
those who have seen its design and operation but their major
concern is low market availability and inaccessibility to fre-
quent maintenance services by the manufacturer.
Vertical mixer
90%
Horizontal
ribbon mixers
4%
Horizontal
Paddle mixers
4%
Other
2%
Fig. 4 Types of mixer designs in the system.
Fig. 5 Department of Agricultural and Biosystems Engineering
(DABE) feed mixer.
Fig. 6 SIS Engineering feed mixer.
Performance evaluation of mechanical feed mixers 4909
3.3. Performance evaluation of mixers
The performance of the feed mixers was mostly linked to the
design of the various important components. Table 2 presents
data on dust pollution associated with the studied feed mixers.
The design of SIS Engineering feed mixer did not include a
cyclone – a device that captures dust produced by the feed mix-
ing machining operations. It was observed in the case of the
SIS Engineering type, maize was milled directly into an open
top hopper and this resulted in dust pollution. The unique fea-
ture of the DABE type is the cyclone at the top of the feed
mixer with hanging dust bags. The dust bags trap the dusts
produced during mixing and milling. The hanging dust bags
make control of dust easy and also facilitate monitoring of
the quantity of dusts. VA Engineering had a vent covered with
sack at the top of the feed mixer to trap dusts. All vertical mix-
ers need a cyclone to separate fine feed particles from exiting
air. The horizontal mixers were mostly airtight, hence no dust
pollution was observed.
The remarks of the various mixer makes’ (Table 3) were the
comments from operators and observations made during the
studies.
DABE mixer has a unique hopper design (Fig. 5) that
allows or gives the operator flexibility to rotate the hopper
about the vertical axis (i.e. around the mixing auger). This
function is to permit or give the operator options to turn the
longer side towards him for ingredient loading and to turn it
towards the ejecting hopper to allow what is called recycling
of feed. The shorter side is turned towards the ejecting spout
during the ejecting of mixed feed. The other unique feature
is the cyclone at the top of the mixer, with the dust bags hang-
ing downwards. This allows trapping of dust produced during
mixing and milling. The hanging dust bags make handling of
dust easy and permit easy monitoring of dust quantity. Gener-
ally, the comment from operators using DABE mixers were
good regarding mixer bearings, since they do not change bear-
ings frequently (approximately 2 years and over before chang-
ing, that is for a very busy mixer). The main flaw observed has
to do with the design of the hammer mill hopper, which is ele-
vated above the ground (about 70 cm), which makes loading
very inconvenient for operators. The overall mixing time is
about 30 min for a 2 tonne mixture. Due to the design of com-
bine hammer mill and mixer, the ground maize has to be air-
lifted into the mixer through a duct.
SIS Engineering mixers uniqueness has to do with its wide
and fixed hopper, and its hammer mill that is installed to mill
directly into the mixer hopper (Fig. 6). This design has the
advantage of short or quick overall mixing time of about
20 min (i.e. milling and mixing time). The hammer mill hopper
is in such a way that it is about one metre above the ground
and makes loading difficult. Operators complain about fre-
quent damage of bearings, which is an evidence of wrong fit
between auger shaft and bearing bore.
VA Engineering mixers stand out with their hammer mill
hopper design; They are very wide and less steep and installed
to flash with the milling room flour (see in Fig. 7). This enables
easy loading of maize, there is no need for lifting, hence drud-
gery during operation is minimal. The hopper is designed wide
and fixed and part is covered with a grate (where ingredient is
fed into the mixer) to prevent sacks from falling into the hop-
per. The major flaw sighted was the angle of repose of the
hammer mill hopper, which was too small, hence maize cannot
flow by gravity and the suction air from the hammer mill could
not move all the maize, unless it is pushed close to the entrance
of the duct.
All of the foreign mills were generally airtight and dust was
not produced during operations. The problem is with the fact
that it is difficult to take samples at various stages during oper-
ation (like after hammer mill) for checks, unless at the end of the
mixing. The SM mixer was unique, since it is designed like a mini
plant (Fig. 9). The entire system comprises a loading hopper,
Fig. 7 VA Engineering feed mixer.
Fig. 8 Foreign make feed mixer from Denmark.
4910 M. Adusei-Bonsu et al.
Fig. 9 Supermach Enterprise feed mill.
Fig. 10 Foreign feed mill from Netherland.
Table 1 Machines performance parameters.
Machines Type of
mixer
Capacity
(t)
Frequency Energy consumption per
mix (kWh)
Speed
(rpm)
Mixing time
(min)
Hammer mill perforated screen
size (mm)
SIS
Engineering
Vertical 1 5 1.9 300–500 15 6–8
2 7 3.1 500 25 6–8
3 1 5.5 500 30 6–8
DABE Vertical 1 1 0.8 350–500 15 5.5–6
2 7 1.9 350–500 15 5–6
Foreign
(FR)
Slanted 0.5 1 0.5 100 15 5.5
Horizontal 1 3 0.9–3.8 40–80 1.5–10 5.5
Supermach Vertical 1 1 0.8 350 15 6.5
VA
Engineering
Vertical 1.5 13 1.9–3.8 250 15–30 6.5–8.5
2 11 2.7–5.4 300 15–30 6.5–8.5
Pf=50
Performance evaluation of mechanical feed mixers 4911
hammer mill feeder auger, ground maize silo, mixer with a
cyclone outside the milling room, mixed feed elevator, and
mixed feed silo. The advantages with the design is that it permits
milling maize into a silo whilst mixing feed at the same time. The
mixed feed is ejected by opening a valve by the side of the mixer
hopper and the entire feed is transferred into another silo for
ejection. The most important thing is there is always continuous
production; no need for pausing to mill maize or waiting to bag
before mixing. The maintenance cost per production capacity is
one of the key factors that determine the effectiveness of the
operation of the feed mixing machine. Table 4 presents the
results of the production capacity and maintenance cost associ-
ated with the various machine makes.
It can be seen from Table 4 that for the entire mixer makes,
the higher the production capacity per day, the higher the aver-
age maintenance cost. The most important concern has to do
with the number of operators. Since DABE and SIS Engineer-
ing makes have their hammer mill above the ground, at least
two persons are needed to load or grind the maize for mixing
(for higher capacities), whilst one person will be loading or
adding the ingredient. Foreign, VA Engineering, and SM do
not have such problems, only two persons can serve as
operators.
3.3.1. Moisture content of feed
The average moisture content of mixed feed per specific mixer
make are listed in Fig. 13. The averages for all the farms were
below 14% (w.b). Feed samples were extracted from 8, 8, 17, 4
and 2 DABE, SIS Engineering, VA Engineering, Foreign
Table 3 Operating performance of key component of feed mixer.
Machine
makes
Machine
component
Remarks
DABE Hammer mill
hopper
dMaize flow freely into the grinding chamber
dMaize has to be lifted (70 cm) before pouring into it
Mixer hopper dIt is flexible (movable), to permit easy ejecting of feed
dIt’s difficult to eject remaining mixed feed in hopper without risk of cut from auger
Mixer bearing dIt lasts longer (more than two years)
Auger and inner
tube
dThe fit between the auger and inner tube is good (5 mm)
SIS
Engineering
Hammer mill
hopper
dMaize flow freely into the grinding chamber
dMaize have to be lifted (1 m) before pouring into it
Mixer hopper dThe mixer hopper is wide and fixed, which makes it easy to add ingredients/concentrate
dMixed feed ejection is quite difficult (spills) and susceptible to cut when bailing out feed left in hopper
Mixer bearing dEasy to replace
dDoes not last (less than three months)
Auger and inner
tube
dThe fit between the auger and inner tube is good (5 mm)
VA
Engineering
Hammer mill
hopper
dEasy to load/feed maize (does not need to lift because hopper flushers with the floor)
dHave to push maize close to inlet duct
Mixer hopper dWide and fixed hopper to allow easy adding of ingredients
dMixed feed ejection is quite difficult (spills) and susceptible to cut when bailing out feed left in hopper
Mixer bearing dIt lasts longer (more than two years)
Auger and inner
tube
dThe clearance between the auger and inner cylinder is >5 mm
Foreign Hammer mill
hopper
dMostly fed by an elevator, which has a loading hopper that flushers with the ground
dExtra electricity cost for elevator motor
Mixer hopper dNo hopper, it is fed by an elevator and the system is airtight
dExtra cost for elevator motor
Mixer bearing dIt lasts more than two years (no feed or debris enters it)
Supermach Hammer mill
hopper
dEasy to load (feeding auger’s hopper flushers with floor)
dExtra cost for auger motor
Mixer hopper dMixer hopper is small and fixed and lower, which makes loading of ingredients easy and has a spout that
links to and an elevator to aid in ejecting feed after mixing
dExtra cost for ejecting elevator motor
Mixer bearing dIt lasts longer (more than two years)
Auger and inner
tube
dThe fit between the auger and inner tube is good (5 mm)
Table 2 Dust pollution associated with feed mixers.
Machine
makes
Component Remarks
SIS
Engineering
Without
cyclone
The mill and feed room were dusty.
DABE With
cyclone
The dust was collected in dust sacks
attached to the cyclone, hence no dust
pollution.
VA
Engineering
Without
cyclone
There was a vent covered with sack
but difficult to remove and makes’ the
mill quite dusty.
Supermach
(SM)
With
cyclone
There was no dust pollution.
Foreign
(FR)
Without
cyclone
The whole system was airtight and
had silos, which stored feed temporary
before ejecting, hence no dust
pollution.
4912 M. Adusei-Bonsu et al.
makes and Supermach mixers, respectively. Three samples
were taking from each feed mixer and the average (M) mois-
ture content calculated. Since the variations in the moisture
content were not vast, the standard deviations (SD) were
small. The moisture content of the mixed feed samples ranged
between 9.21% and 11.08% as shown in Fig. 13. The values
were within the safe moisture content range for mixing, which
is 14% (Amplifies Ghana, 2017). The chain line in Fig. 13
shows the safe moisture content boundary; above which mix-
ing is not effective. Thus, since all the moisture contents were
below 14%, it was good for mixing.
3.3.2. Feed particle size distribution
The particle size distribution for the various mixer makes are
shown in Figs. 14–18 and summarised in Table 5. The mean
particle sizes of samples taken from different farms mixed by
the same mixer makes were not significantly different (Table 5).
The samples were extracted from eight (8) DABE feed mixers
with three (3) replicates for this experiment and the average
presented in Fig. 14. The particle sizes determined from the
experiment were a bit widely spread from the average, hence
relatively high standard deviations. This variation in particles
is due to the fact that the various mills use different hammer
mill screen sizes.
Samples were extracted from eight SIS engineering mixers
with three replications. The average results from the experi-
ment are presented in Fig. 15. The various mills used different
hammer mill screen sizes, hence higher SD values due to the
spread of values from the average.
Nineteen VA engineering mixers were used in the particle
size distribution experiment. It is important to note that the
hammer mill screen sizes for these mills varied, hence widely
spread values from the average, hence given rise to higher
SD values as shown in Fig. 16.
The standard deviation values presented in Fig. 17 are rel-
atively high; indicating greater variation in the values. This
was due to the fact that the various feed mixers used different
hammer mill screen sizes.
Samples were extracted from two Supermach mixers for the
experiment at three replications. SD values indicate that
the results were widely spread from the average values and
are due to different hammer mill screen sizes used by the two
mills.
From Figs. 14–18, the particle uniformity coefficients were
read and calculated, and presented in Table 5. From Table 5,
the fineness modulus (FM) of all the samples mixed by the dif-
ferent mixers ranged from 3.11 to 3.42, which is within the
Table 4 Production capacity of machines and maintenance cost.
Machine make Type of mixer Capacity (t/d) Number of operators Average maintenancecost (GhC) Average maintenancecost (US$)
DABE Vertical 2 2 173 38.19
>2–4 2 200 44.15
>4–10 3 200 44.15
>10 4 210 46.36
SIS Engineering Vertical 2 3 55.6 12.27
>2–4 3 200 44.15
>4–10 4 240 52.98
>10 4 400 88.30
VA Engineering Vertical 2 2 30 6.62
>2–4 2 124 27.37
>4–10 2 130 28.70
>10 2 135 29.80
Foreign Slanted <2 2 70 15.45
Horizontal >10 2 500 110.38
Supermach Vertical >10 3 100 22.08
SIS
Engineering
54%
DABE
29%
Foreign
17%
Fig. 11 Percentage availability of feed mixers in Ashanti region.
VA
Engineering
88%
Supermach
4%
DABE
4%
Foreign
4%
Fig. 12 Percentage availability of feed mixers in Brong Ahafo
region.
Performance evaluation of mechanical feed mixers 4913
standard fine particles (uniform) range of 2.0–4.0 FM that is
good for mixing [16].
The average particle size or geometric mean diameter
(GMD) of the feed samples mixed by all the mixer makes ran-
ged from 762.74 to 947.27 mm with geometric standard devia-
tion (GSD) range of 2.35–2.50. Thus having some of the
average particle size within the target mean particle size range
of 600–900 mm for mash diet for poultry and also all GMD
were less than the range of particle size (1200–1500 mm) that
makes micro-ingredient incorporation difficult during mixing
[19,20].
The particle size distribution used for the feed sample
preparation were diverse. For the DABE mixers (Fig. 14),
about 32% of particles were coarser than 710 mm composition
and for SIS Engineering, VA Engineering, Foreign and SM
(Figs. 15–18), about 36%, 37%, 38% and 42% materials were
coarser than 710 mm, respectively. The particle size distribu-
tions for all the feed samples from the various mixers had
greater percentage of their particle sizes less than 710 mm,
which is also indicated by the uniformity index (UI) values
(Table 5). All UI values for the different mixer makes had most
of their particle size within the medium and fine zone (less than
one millimetre). The UI values are within standards for layers
at their production/commercial stage [21].
According to Herrman and Behnke [19], size uniformity of
the finished feed ingredients can impact directly on the final
ingredient dispersion. When all the physical properties of the
constituents are relatively the same or within permissible
range, then mixing becomes fairly simple. Hence the quality
of blending or how well the various ingredients are mixed, is
entirely dependent on the mixer structures and time of mixing.
Item
M ± SD
n =
Feed from
DABE
10.945 ± 2.0883
24
Feed from
SIS
9.2125 ± 0.8788
24
Feed from
VA
10.1584 ± 1.0423
57
Feed from
Foreign
10.8317 ± 0.8380
12
Feed from
SM
9.46 ± 0.4503
6
Fig. 13 Average moisture content of sampled feed from the various feed mixers.
Sieve
M ± SD
n
8000
0.0136 ± 0.0766
24
4000
1.3591 ± 2.0869
2000
19.4513 ± 8.2082
1000
31.1546 ± 3.6655
500
23.4691 ± 5.7728
250
12.2617 ± 3.1400
125
10.0700 ± 3.9378
Pan
2.2206 ± 0.2697
Fig. 14 Particle size distribution of sampled feed from DABE mixers.
Sieve
M ± S
n
8000
0.0350 ± 0.1212
24
4000
2.0336 ± 2.8860
2000
21.2756 ± 9.9918
1000
29.0792 ± 3.5007
500
23.2506 ±7.5186
250
12.5634 ± 3.4451
125
9.5115 ± 4.0974
Pan
2.2512 ± 1.7777
Fig. 15 Particle size distribution of sampled feed from SIS Engineering mixers.
4914 M. Adusei-Bonsu et al.
Sieve
M ± SD
n
8000
0.0327 ± 0.1902
12
4000
1.6593 ± 1.6881
2000
20.1026 ± 7.1754
1000
29.8237 ± 3.8264
500
24.1418 ± 4.5755
250
12.1270 ± 3.0263
125
10.4186 ± 4.2136
Pan
1.6942 ± 1.8294
Fig. 16 Particle size distribution of sampled feed from VA Engineering mixers.
Sieve
M ± SD
n
8000
0.0230 ± 0.0990
12
4000
1.5145 ± 2.4638
2000
18.8937 ± 9.0999
1000
30.7709 ± 3.8063
500
23.9790 ± 6.3582
250
12.4891 ± 3.1897
125
10.6803 ± 3.9396
Pan
1.6497 ± 1.6911
Fig. 17 Particle size distribution of sampled feed from Foreign mixers.
Sieve
M ± SD
n
8000
0.0124 ± 0.0731
6
4000
1.2969 ± 1.9979
2000
18.9949 ± 7.9627
1000
31.0031 ± 3.5442
500
23.516 ± 5.5085
250
12.3235 ± 3.0769
125
10.6528 ± 4.2017
Pan
2.2047 ± 2.0586
Fig. 18 Particle size distribution of sampled feed from Supermach mixer (SM).
Table 5 Feed characteristics determination parameters from various types of machines.
Machine make from which feed were
sampled
Fineness
modulus
Geometric mean
diameter(mm)
Geometric standard
deviation
Uniformity index (Coarse:
Medium: Fine)
DABE 3.36 913.43 2.35 2.1 : 5.4 : 2.4
SIS Engineering 3.41 940.16 2.42 2.4 : 5.2 : 2.4
VA Engineering 3.42 947.27 2.43 2.3 : 5.3 : 2.3
SM 3.11 762.74 2.50 1.5 : 5.3 : 3.2
Foreign 3.25 839 2.36 1.5 : 5.9 : 2.6
Performance evaluation of mechanical feed mixers 4915
3.3.3. Blending uniformity test
Table 6 presents the averages and coefficient of variations in
percent salt of sampled feed from the various mixer makes.
The standard values and the respective corrective actions are
also presented. Two separate foreign mixers, and four local
ones; VA Engineering mixer, SIS Engineering mixer, DABE
mixer, and SM mixer were used for the blending uniformity
test.
Percentage CV of measured quantity of salt (0.3% of the
ingredient weight) in mixed feed sample is an internationally
recognized method of determining uniformity of mixing. The
recommended industrial value is 10% and below, which is
the percent commonly recognized by the feed industry as the
cut-off for uniformity of mix analysis [22]. Herrman and
Behnke [19] had provided a chart of standard CV value ranges,
ratings and their corrective action to guide in determining the
performance of mixers. When the CV of a mixer is above 10%,
it means that the mixer is not able to intersperse the ingredients
uniformly given its working condition and all other factors (i.e.
particle size distribution, mixing time, moisture content, etc.).
The CV for DABE, SIS, and SM mixers in Table 6 suggest
their mixing/blending uniformity is good. Using Herrman and
Behnke [19] chart, this means all these mixers have no inner
structural deformities but need to increase their mixing time
of 15 min by 25–30%. Although the three mixer makes are
in the same performance zone, their CV were 10.30%,
10.80% and 10.93% for SIS, DABE and SM, respectively.
With CV values of 18.72% and 18.05% for the foreign
mixer (from Denmark) and VA Engineering respectively, the
feed mixers performed fairly [19]. These values suggest that
the two feed mixer types might have had some defective parts.
In the case of the foreign feed mixer from Denmark, it was
quite old (installed in the early 19900s) and upon inspection
there was evidence of worn out screw auger and the main gear
drive. VA Engineering mixers were relatively new (within
5 years after installation) and with the two mixers that were
selected for the test, there was no evidence of worn out auger
or tube. According to Wicker and Poole [23], when the screw
diameter of a mixer reduces by 12.5 mm, mixing time should
be increased by 5 min. This is because the standard maximum
clearance of 5 mm between the inner tube and the auger is
exceeded, resulting in low mixing efficiency. In the case of
VA Engineering feed mixer type, the rating suggests worn
out component, hence component designs should be checked.
Essential components such as the auger and the inner tube,
and the gap between inner tube and the base of the cone (bot-
tom mixing area) should be 500–800 mm. In order to achieve
uniform mixing for both the VA Engineering type and the type
from Denmark, mixing time has to be extended more than
three (3) times the current time, which will not be energy effi-
cient, and hence cost effective.
For all the feed mixers that were tested for blending unifor-
mity, the foreign feed mixer from Netherlands indicated a
CV = 7.19%, which was less than the recommended industrial
value of 10% [19]. This means the feed mixer probably had no
worn out parts, hence the mixing time of 1.5 min is adequate
for efficient and effective mixing. The results in Table 6 indi-
cates that increasing mixing uniformity (mixing efficiency)
can result in reducing mixing time which makes mixing cost-
efficient [24].
3.4. Summary of findings
From the study results, all the major determining factors of
efficient blending of feed ingredient relating to the feed ingre-
dients characteristics were standard or within permissible
range. The performance of the feed mixers therefore is depen-
dent on the design whether it is user-friendly, low maintenance
cost, environmentally friendly etc. User-friendly designs have
direct relationship with the number of operatives required to
operate the machine. In general, user friendly designs of
machines take into consideration ergonomics, particularly
motion, posture, load lifting etc. that should fit the physical,
physiological and sociological needs of the operatives. Tables
3 and 4 indicate that the higher the hammer mill hopper above
the ground, the more the number of operators needed. In
terms of hammer mill hopper designs, the foreign mixers,
VA Engineering, and Supermach mixers were found to be
Table 6 Coefficient of variation of NaCl in feed sampled from the various mixers.
Mixer make Means of %
NaCl in feed
sample
Standard deviation
of % NaCl in feed
sample
Coefficient
of variation
(%)
Standard
%CV
Rating Corrective action
DABE 0.966 0.104 10.80 10–15% Good Increase original mixing time by 25
to 30%
SIS 0.875 0.090 10.30 10–15% Good Increase original mixing time by 25
to 30%
VA 0.759 0.137 18.05 15–20% Fair Check and replace worn equipment,
overfilling, or sequence of
ingredient addition
Foreign make from
Denmark (App. A,
Fig. 4)
0.851 0.159 18.72 15–20% Fair Check and replace worn equipment,
overfilling, or sequence of
ingredient addition
Foreign make from
Nertherland (Used by
Agricare Ltd.)
0.85 0.061 7.19 10% Excellent None
SM 0.938 0.103 10.93 10–15% Good Increase original mixing time by 25
to 30%
Standard %CV, rating and corrective action in table (Source: [19;17]).
4916 M. Adusei-Bonsu et al.
ergonomically better because the height of the hoppers were
relatively low, making them easy to use.
Maintenance cost per month is an indication of how fre-
quent the machine develops fault and the severity of the faults.
The results in Table 4, the maintenance cost of the foreign type
feed mixers with higher production capacity of about 10 ton-
nes/day was about Ghȼ500 (US$ 110) per month. Feed mixers
whose monthly maintenance cost were relatively low included
VA Engineering Ghȼ30 (US$ 6.6), SIS Engineering 2 tonnes -
Ghȼ55.6 (US$ 12.3), and Supermach Ghȼ100 (US$22). Main-
tenance cost per month of DABE Ghȼ173–210 (US$ 38–46)
and SIS Engineering- Ghȼ200–400 (US$ 44–88) per month.
It is agreed that the simpler the design of a machine, the less
component parts it has, hence less maintenance cost.
All the foreign mixers produced less dust because they were
air-tight. Those with pneumatic conveyors also had cyclones to
separate the fine particles from the exhaling air. For the local
mixers Supermach and DABE dust production into the envi-
ronment was minimal, because a dust collecting cyclone was
incorporated in the design. But Supermach mixer was much
safer during discharge of mixed feed, because it does not
require manual bailing-out of remained feed in the mixer hop-
per, unlike the other local makes. In terms of mixing unifor-
mity, the foreign mixer (horizontal paddle type mixer)
indicated better value of CV = 7.19%. Three of the locally
manufactured feed mixers (SIS, DABE and Supermach) whose
performance metrics were relatively low would need to opti-
mize their mixing time to be below the standard cut-off for uni-
formity of mix analysis in the feed industry.
4. Conclusions
The performance of four local feed mixing machines, which
were all vertical types and two foreign makes were evaluated
based on their machine parameters, operational parameters
and feed characteristics. The mixing capacities of both local
and foreign makes were similar in tonnage. With relatively
high mixing speed and mixing time, the vertical feed mixers
consumed more energy per mix in kWh than the horizontal
and the slanted feed mixers. Dust collection cyclone is a key
component of the feed mixing machine that helps to capture
dust during the machining operations. From the study, the ver-
tical feed mixers without cyclones had dust pollution issues,
whereas the horizontal and the vertical types with cyclones
were better at managing dust pollution. In regard to the feed
characteristics, blending uniformity of three vertical feed mix-
ers were judged as good with coefficient of variation values a
little above the industrial standard value. This implies that
their mixing times would require adjustments to attain the
industrial standard coefficient of variation value or below.
The blending uniformity of the slanted feed mixer was fair
and that was due to worn parts. The horizontal mixer type
had an excellent blending uniformity with a coefficient of vari-
ation value below the industrial standard value. It is recom-
mended that blending uniformity test, at specific time
intervals should be carried out by local manufacturers of feed
mixers to determine the optimum time to achieve better mixing
(CV less than 10%) before delivering feed mixers to end users
of poultry and feed industries. Considering all the performance
metrics that were used for the analysis, the horizontal paddle
type mixer indicated better metrics than the locally manufac-
tured feed mixers of Supermach, followed by DABE and SIS
Engineering. Given that the horizontal feed mixers were found
to have better performance metrics, there is the need to adapt
the design for improved performance, while focusing on substi-
tuting imported machine parts and components with locally
available materials to minimise maintenance cost for the poul-
try and feed industries. VA Engineering feed mixer, in partic-
ular, should redesign some components such as hopper, auger
and inner cylinder to enhance its performance.
Declaration of Competing Interest
The authors declare that they have no known competing
financial interests or personal relationships that could have
appeared to influence the work reported in this paper.
References
[1] FAO, Poultry in the 21st Century: avian influenza and beyond,
Proceedings of the International Poultry Conference, held 5–7
November 2007, Bangkok, Thailand. Edited by O. Thieme and
D. Pilling. Food and Agricultural Organisation (FAO) Animal
Production and Health Proceedings, No. 9. Rome, 2008.
[2] V. Ravindran, Nutrition and Human Health, Monogastric
research centre, Institute of Food, Massey University,
Palmerston North, 2013.
[3] A. Mitchell, Global poultry trends - Developing Countries Main
Drivers in Chicken Consumption, The Poultry Site, Available
online, [Accessed 24/06/2020], https://thepoultrysite.com/
articles/global-poultry-trends-developing-countries-main-
drivers-in-chicken-consumption, 2016
[4] T.L. George, Composition of mixing system, McGraw Hill
Book Company, London, 2009.
[5] C. Narrod, C. Pray, Technology transfer, policies, and the
global livestock revolution, Proceedings of the International
Agricultural Trade Research Consortium Symposium on Trade
in Livestock Products, held 18–19 January 2001, Auckland, New
Zealand, 2001
[6] C. Narrod, K. Fuglie, Private-sector investment in livestock
breeding, Agribusiness Int. J. 4 (6) (2000) 457–470.
[7] S.O. Onyegu, O. Ibifuro, M.N. Idung, T. Aguheva, Design and
fabrication of an industrial poultry feed tumble mixer, Leonardo
Electron. J. Practices and Technol. 20 (2012) 49–60.
[8] C. Culpin, Farm Machinery, 11th Edn., Mackays of Chathan,
Kent, 2000.
[9] A.A. Balami, D. Adgidzi, A. Mua’zu, Development and testing
of an animal feed mixing machine, Int. J. Basic Appl. Sci., 01(3)
(2013) 491-503.
[10] M.B. New, Feed and feeding of fish and shrimp, ADCP/REP/
87/26, FAO/UNDP, Rome, 1987.
[11] C.G. Scanes, G. Brant, M.E. Ensminger, Poultry science. Upper
Saddle River, New Jersey, USA, Pearson Prentice Hall, 2004.
[12] Y. Trillo, A. Lago, N. Silva-del-Rı
´o, Total mixed ration recipe
preparation and feeding times for high-milk-yield cows on
California dairies, The Professional Animal Scientist 33 (4)
(2017) 401–408.
[13] J.G. Brennan, J.R. Butters, N.D. Cowell, A.E.V. Lilley, Food
Engineering Operations, Elsevier Applied Science 3 (1998), 91–
107 and 287–289.
[14] A. Addo, A. Bart-Plange, Performance evaluation of two on-
farm feed mixers, Agricultural Mechanization in Asia, Africa,
and Latin America 39 (3) (2008) 57–60.
[15] ASABE Standards, Densified products for bulk handling—
definitions and method, ASABE S269.5, in: ASABE Standards
2012, American Society of Agricultural and Biological
Engineers, St. Joseph, MI, 2012, p 1–8
Performance evaluation of mechanical feed mixers 4917
[16] ASAE, ‘Method of determining and Expressing fineness of feed
materials by sieving’, ANSI/ASAE S319.3 FEB03, American
Society of Agricultural Engineers (ASAE), Michigan, 2005
[17] C. Stark, M. Saensukjaroenphon, Testing Mixer Performance,
Kansas State University, K-State Research and Estension
Service, Manhattan, 2017.
[18] Feedmachinery, Ensuring Optimum Mixability In Feed
Manufacturing: Part 3 - Factors That Affect Mixer Performance,
https://www.feedmachinery.com/articles/feed_technology/
mixability3/; (accessed March 2021), 2021
[19] T. Herman, K. Behnke, Testing mixer performance, MF-1172.
Kansas State University Agricultural Experiment Station and
Cooperative Extension Service Bulletin. Kansas State
University, Manhattan, 1994
[20] D. Ziggers, Correct particle size benefits animal performance,
Feed Tech 13 (5) (2009) 2009 (accessed March 2021) https://
www.allaboutfeed.net/home/correct-particle-size-benefits-animal-
performance/,.
[21] Hy-Line International, Feed granulomere and the importance
of feed particle size in layers, Technical update (2016).
[22] A. Fahrenholz, C.R. Stark, Mixing feeds and mixer test
procedures for batch mixers, in: T. Lundeen, M.N. Minnetonka
(Eds.), Feed additive compendium. Miller Publishing Co, 2014,
105-108.
[23] D.L. Wicker, D.R. Poole, How well is your mixer performing?,
Feed Manage. 42 (9) (1991) 40–44.
[24] A. Farahinia, J. Jamaati, H. Niazmand, W.J. Zhang, Numerical
investigation into the mixing performance of micro T-mixers
with different patterns of obstacles, J. Braz. Soc. Mech. Sci. Eng.
41 (11) (2019) 491.
4918 M. Adusei-Bonsu et al.
