Agronomy Research 13(1), 89–94, 2015

Continuous airflow rate control in a recirculating batch

grain dryer
T. Jokiniemi*, T. Oksanen and J. Ahokas

University of Helsinki, Faculty of Agriculture and Forestry, Department of Agricultural

Sciences, P.O. Box 28, 00014, University of Helsinki, Finland;
*
Correspondence: tapani.jokiniemi@helsinki.fi

Abstract. As the energy efficiency requirements in agriculture increase, offers grain drying
opportunities for substantial energy saving. Earlier work indicated that energy savings in grain
drying can be achieved by controlling the drying airflow rate during the drying process. Aim of
this study was to design an embedded control system, based on microcontroller, for continuous
airflow rate control in a recirculating batch grain dryer, and to test it in a scaled-down research
dryer. The control system proved to be working as designed, reducing the dryer airflow rate
smoothly towards the end of the process. However, additional research of the energy efficiency
and performance of the dryer using the airflow rate control is needed.

Key words: grain drying, grain dryer, energy efficiency, airflow rate control, drying air
temperature.

INTRODUCTION

Grain drying is one of the largest energy inputs in arable farming in boreal and
northern temperate climate zone countries (Mikkola & Ahokas, 2009). Considering the
ever increasing energy efficiency requirements in agriculture, provides grain drying, and
grain preservation in general, opportunities for substantial energy savings. Previous
work of the authors indicated that controlling the airflow rate and drying air temperature
during the drying process has potential to improve both the energy efficiency and the
performance of the dryer (Jokiniemi & Ahokas, 2014). The energy savings were
achieved by decreasing the dryer airflow rate towards the end of the process and
elevating the drying air temperature concurrently. The energy savings were based on two
factors: 1) using higher drying air temperature and 2) obtaining higher exhaust air
humidity in the latter part of the drying process.
The effect of the elevated drying air temperature on the energy efficiency of the
dryer has been reported by several authors (Morey, Cloud & Lueschen, 1976; Suomi
et al., 2003). This is a consequence of the nonlinearity in the moist air equilibrium
equations; the same amount of added heat energy increases the water binding capacity
of air more in higher temperatures, as Fig. 1 indicates. Additionally, the elevated drying
air temperature creates a greater pressure gradient of water vapor between the core and
the surface of the whole grains, which enhances the water movement inside the grain
and hence also the evaporation.

89
 In addition to the drying air temperature, the humidity of the exhaust air has a
significant effect on the energy efficiency of a dryer, as shown by Fig. 1. While the
heating power of the dryer remains usually constant, the relative humidity of exhaust air
indicates directly the efficiency of evaporation. When the exhaust air humidity is high,
the majority of the applied heat energy appears as latent heat in the exhaust air, and the
exhaust air temperature is typically low. When the relative humidity of the exhaust air is
low, the applied heat appears as sensible heat. It is not thus used for evaporation, but it
is lost with the exhaust air. In this case the exhaust air temperature is relatively high and
the specific energy consumption in drying is high. When the airflow rate is reduced, the
moisture in the grain has more time to diffuse to the surface of individual whole grains.
Higher exhaust air relative humidity is thus obtained, leading to lower energy
consumption in drying. The effect of the airflow rate on the energy consumption in grain
drying has also been recognized by several authors in past. For example Morey et al.
(1976) and Peltola (1988) suggested airflow rate control as one possible approach for
reducing the energy consumption in grain drying.

Figure. 1. The effect of drying air temperature T and exhaust air relative humidity RH on the
energy consumption of adiabatic drying process (on the left) and typical exhaust air properties in
practical grain dryer (on the right).

Another requirement for the control of grain dryer is to maintain the quality
properties of grain, such as germination ability and baking qualities. Previous studies
have proved that using high drying air temperature may severely damage these
properties, but this vulnerability is also strongly affected by the moisture content of the
grain. The grain can tolerate higher temperatures when its moisture content is reduced
(Ghaly & Taylor, 1982; Ambardekar & Siebenmorgen, 2012). Therefore it is reasonable
to assume that the grain is not so sensitive to heat damage in the latter part of the drying
process, when some drying has already occurred, and the drying air temperature can be
increased gradually. This is also a beneficial approach considering the energy use, since
the dryer exhaust air humidity is inherently high in the beginning of drying, leading to
efficient energy use, and begins to decrease towards the end of the process (Jokiniemi &
Ahokas, 2014). The benefits of the reduced airflow rate and elevated drying air
temperature, considering the energy use, are thus essential in the latter part of the
process, which can also be concluded from the Fig. 1.
In the previous work of the authors, the airflow control was conducted by manual
control of the speed of the drying air fans when the exhaust air humidity had decreased

90
under a certain level. It was concluded that an automatic control system, which utilizes
the exhaust air humidity and temperature information as control factors, would be
relatively simple and easy to implement in new or existing grain dryers. (Jokiniemi &
Ahokas, 2014) Aim of this work was to build a simple microcontroller based embedded
control system for the continuous airflow control in a recirculating batch grain dryer,
and test it in a scaled-down research dryer.

MATERIALS AND METHODS

The study was conducted in a scaled-down research dryer at the research farm of
the University of Helsinki. A measuring system for observing the essential process
variables, such as air temperatures and humidities as well as the air flow rate, was
installed into the dryer. A detailed description of the research dryer and the measurement
system, including the used sensors and data acquisition, can be found in the previous
work of the authors (Jokiniemi & Ahokas, 2014). The measuring system enabled the
calculation of the essential process parameters, such as the amount of removed water
and energy use in a point in time, as well as during the entire process in total. The amount
of removed water was calculated from the changes in the air humidity, while it passed
through the grain, and the air flow rate. The heat power was calculated from the changes
in the enthalpy of the dryer supply air before and after the heaters, and the air flow rate.
Energy consumption was calculated by multiplying the heat power with the measuring
interval, and total energy consumption during each test run was received by summing
the energy consumptions recorded for all of the measuring intervals.
The airflow rate in the research dryer was controlled by a frequency converter,
which was used to control the speed of the drying air fans. The frequency converter could
be controlled either manually from the control panel, or by using a 0–10 V external
control signal. The 0–10 V input was used to supply the control signal from the
microcontroller to the frequency converter.
The airflow rate controller was based on an Arduino Mega 2560 microcontroller
development board. A simple setup of two Honeywell HIH-4000 humidity sensors and
one LM35 temperature sensor was configured to the Arduino board, and a pulse width
modulation (PWM) signal from the microcontroller was used as the control signal to the
frequency converter. A control algorithm to generate the PWM-signal on the basis of the
voltage readings from the humidity sensors was written in C-language and uploaded to
the microcontroller. While the voltage range expressed by the PWM-signal from the
microcontroller digital pins was 0–5 V, a power transistor circuit was built to repeat the
PWM-signal with a higher voltage level. A supply voltage of 12V was used, and suitable
coefficients in the microcontroller program were used to scale the control signal to the
correct range for the frequency converter.
The aim of the control algorithm was to decrease the speed of the drying air fans
smoothly when the humidity of the exhaust air started to decrease. The control rules were
defined by Eq. (1):

‫ݑ‬ோு ሺ‫ݐ‬ሻ ൌ ‫ݑ‬௜௡௜௧ െ ሺܴ‫ܪ‬௟௜௠௜௧ െ ܴ‫ܪ‬ሺ‫ݐ‬ሻሻ ൉ ‫ܭ‬ (1)

where: uRH controller output; uinit = constant term; RHlimit = controller threshold RH;
RH = measured RH; K = controller gain.

91
 The threshold value RHlimit for the RH of the exhaust air was set to 90%, i.e. the
controller became active when the RH of the exhaust air reduced below this level. The
constant term uinit was the desired control value in the beginning of the drying process.
The control algorithm was thus almost equal to the conventional P-controller. However,
due to the nature of the process, the RH of the exhaust air decreases inevitably towards
the end of the process, and the control algorithm will thus end up reducing the output
while the process proceeds. Additional conditional expression was added to the
microcontroller program to ensure that the control output uRH did not exceed the constant
term value uinit in the beginning of the process, when the RH was greater than 90%.
The drying air temperature was adjusted manually to 65 °C in the beginning of the
drying process, and it was allowed to rise freely as the airflow was decreased. However,
an upper limit of 90 °C was defined for the drying air temperature to avoid heat damage
to the grain. The microcontroller program compared the temperature of the drying air to
the upper temperature limit threshold value, and when it was reached, the temperature
control rule became active:

‫ ் ݑ‬ሺ‫ݐ‬ሻ ൌ ‫ݑ‬ோு ሺ‫ݐ‬ሻ െ ሺܶሺ‫ݐ‬ሻ െ ܶ௟௜௠௜௧ ሻ ൉ ‫ܭ‬ (2)

where: uT = controller output; uRH = output from the RH controller; T = measured drying
air temperature; Tlimit = controller threshold temperature; K = controller gain.

The controller gain value K for both controllers was defined in the first test runs.
The K values used in the final setup were 1 for the RH-controller and 5 for the
temperature controller. The aim for the RH-controller was to decrease the airflow
smoothly in such a way that the upper temperature limit was not reached until close to
the end of the process. For the temperature controller the aim was to react quickly when
the temperature limit was reached, but to avoid the excessive fluctuation of the controller
output caused by too high gain value.
Control system was tested in drying trials in the scaled-down research dryer.
Altogether eight drying trials were accomplished: four with the airflow rate control
system and four references without it. The airflow rate in the reference trials was equal
to the initial airflow rate in the control system trials. The grain used in the trials was
barley with the initial moisture content of ca. 20% (wet mass basis). The drying process
was ended when the moisture content of the barley had decreased to ca. 13.5% in each
trial.

RESULTS AND DISCUSSION

Fig. 2 presents the operation of the controller during the drying process. In the
beginning of the process the exhaust air humidity was high, and the initial set point for
the controller output was used. When the decreasing exhaust air RH reached the
threshold value of 90%, the controller started working. The controller operated as
designed, reducing smoothly the airflow rate and increasing drying air temperature
simultaneously. When the maximum drying air temperature limit was reached at 90 °C,
the temperature controller in Eq. (2) became active, starting to increase the airflow. Some
oscillation occurred in the control signal value in the end of the process, when the two

92
control algorithms were alternating. This did not, however, have any significant practical
effect on the drying air temperature or the airflow rate.

Figure 2. Dryer exhaust air humidity, control signal value and the effect of the control system on
the drying air temperature (on the left) and dryer airflow rate (on the right).

Fig. 3 presents the specific energy consumption and the evaporation rate in each
trial member. The results do not indicate a clear advantage for the airflow rate control
considering the energy efficiency. The average energy consumption was 10% lower with
the controller, compared to the conventional drying. However, the coefficient of
variation for specific energy consumption in the airflow rate control trials was 17%,
while the corresponding figure in the conventional process trials was only 9%.

Figure 3. Average specific energy consumption and evaporation rates in each trial with airflow
rate control and conventional drying without airflow rate control.

The relatively large variation in the airflow rate control trials is explained mainly
by the third test run, which had an exceptionally low energy consumption compared to
the other trials in the Fig. 3. This may be a consequence of a possible measurement error
in this trial.
The evaporation rate for the airflow rate control trials was in average 15% higher
compared to the conventional process, which indicates that the airflow rate control could
be used to enhance also the performance of the dryer. However, the variation was large

93
also here, and the last trial in the conventional process showed an exceptionally high
evaporation rate compared to the others.

CONCLUSIONS

The simple microcontroller based control system for controlling the drying airflow
rate in a recirculating batch grain dryer operated as designed, reducing smoothly the
airflow rate towards the end of the drying process. The results indicated that the control
system enhanced both the energy efficiency and the performance of the dryer, but the
variation in the results was also large. The simple and inexpensive control system could
be easily installed in most of the dryers. In the current study the airflow rate was
controlled by adjusting the speed of the drying air fans by a frequency converter, but a
choke valve in the dryer supply air intake pipe could be used as well. The exhaust air
temperature information could be used as a control input, instead of relative humidity,
as there is a strong inverse correlation between the exhaust air humidity and temperature.
Replacing the sensitive humidity sensors by robust temperature sensors would improve
the reliability of the system considerably. However, further research about the effect of
the control system on the energy efficiency and performance of the dryer is needed prior
to the commercial applications.

ACKNOWLEDGEMENTS. This study was funded by the foundation of Marjatta and Eino Kolli.

REFERENCES

Ambardekar, A.A. & Siebenmorgen, T.J. 2012. Effects of postharvest elevated-temperature

exposure on rice quality and functionality. Cereal Chemistry 89(2), 109–116.
Ghaly, T.F. & Taylor, P.A. 1982. Quality effects of heat treatment of two wheat varieties. Journal
of agricultural engineering research 27, 227–234.
Jokiniemi, H.T. & Ahokas, J.M. 2014. Drying process optimisation in a mixed flow batch grain
dryer. Biosystems engineering 121, 209–220.
Mikkola, H. & Ahokas, J. 2009. Energy ratios in Finnish agricultural production. Agricultural
and food science 18, 332–346.
Morey, R.V., Cloud, H.A., & Lueschen, W.E. 1976. Practices for the efficient utilization of
energy for drying corn. Transactions of the ASAE 19, 151–155.
Peltola, A. 1988. Viljankuivurin säätö kuivauksen aikana – hyödyksi vai hienostelua?
Työtehoseuran maataloustiedote, 355, 8 pp. (in Finnish)
Suomi, P., Lötjonen, T., Mikkola, H., Kirkkari, A.-M. & Palva, R. (2003). Viljan korjuu ja
varastointi laajenevalla viljatilalla. Maa- ja elintarviketalous 31, 100 pp. (in Finnish)

94