IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS

DC motor simulation transfer function estimation: case study Proteus

Ver. 7
To cite this article: S A Wahyu and D P Riky 2019 IOP Conf. Ser.: Mater. Sci. Eng. 674 012040

View the article online for updates and enhancements.

This content was downloaded from IP address 181.215.88.72 on 14/11/2019 at 17:14

The 2nd International Conference on Engineering and Applied Technology IOP Publishing
IOP Conf. Series: Materials Science and Engineering 674 (2019) 012040 doi:10.1088/1757-899X/674/1/012040

DC motor simulation transfer function estimation: case study

Proteus Ver. 7

S A Wahyu and D P Riky

Electrical Engineering. Ahmad Dahlan University, Yogyakarta, Indonesia

E-mail: [email protected], [email protected]

Abstract: Computing technology development today has a significant influence on the design
process of a product or system even including the learning process in the field concerned. The
electronics and control field is a field that greatly helped by these developments. Developments
of electronic products or systems become faster and more efficient with the availability of
CAD (Computer-Aided Design) software. Some Electronic CAD software provides DC motor
model, as Proteus 7.0. Availability of this model is beneficial in control system designing and
learning but the availability of a DC motor models are often constrained because are not
accompanied with adequate information about internal parameters of the dc motor model or dc
motor modeling done in programming so it is not possible to derive dc motor transfer function
based on internal parameters. This study successfully showed transfer function derivation of dc
motor in Proteus CAD simulation with a black-box approach for linear systems.

1. Introduction
Understanding the character of a system is a prerequisite for the analysis of systems[1,7]. Often this
knowledge is also needed in designing a new system composed of several well-known systems before.
For systems that are known in detail we can derive system model from the system components
mathematics model that has been developed previously For a system with unknown subsystem, then
the system identification can be done by observing the system's response to a particular input. This
approach is often referred to as a black box approach [2,3].
Within time domain, the black-box approach includes applied several alternative signals into the
system; some of the signals are step signal, ramp signal, and impulse signal.
Proposals to use the unit step system response for the system transfer function determination started
by Kupmuller [3] and this method is commonly used in industry [4,9].
In this study, we use the system response of step signal input to estimate system transfer function
using MatLab identification toolbox. The system in this study is a dc motor model in Proteus ECAD
tools ver7. A comparison between estimated transfer functions with real transfer function is carried out
to judge validity of estimated transfer function. The contribution of this study is that it allows
performing dc motor modeling on ECAD tools without having to obtain prior internal parameters of
the dc motor.

2. Method

2.1 DC motor modeling

For a dc motor open-loop dynamic equation is[7,8,9] :

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
The 2nd International Conference on Engineering and Applied Technology IOP Publishing
IOP Conf. Series: Materials Science and Engineering 674 (2019) 012040 doi:10.1088/1757-899X/674/1/012040

di
L  iR  k   V (1)
dt b
d
J k i T ( 2)
dt t d
where L is armature inductance (h), R is armature resistance (ohm), V is input voltage (volt), I for
armature current (ampere), kb for back emf constants, and Kt reveals as torque constants.

2.2 DC motor model in ECAD Proteus ver 7.0

Proteus ver 7.0 has DC motor model. DC motor model accommodates dc motor with inertia, and
loading while the dc motor model with encoder is a model that accommodates dc motor and load
inertia and equipped with encoder position. This study used a model of dc motors. The schematic
model for dc motors as shown in figure 1.

Figure 1. DC motor schematic model in Proteus [10]

2.3 DC motor model used in this study

In this study, DC motor with encoder is used. The motor model parameters are shown in Table 1

Table 1. DC motor model parameter being studied

Parameters Nominal
Ra ,(Armature Resistance) 22 ohm
La (Armatur Inductance) 100 mH
Nominal Voltage 5V

The output of Proteus dc motor model is in the form of pulses. These pulses are generated by an
internal rotary encoder. Pulse duty cycle and frequency can be set. Due to output is not in the voltage
form, but in form of pulse, it is necessary to use a frequency-to-voltage converter circuit so that the
block diagram of the system shown in Figure 2.

Step input DC motor simulation F/V converter circuit output

Figure 2. System experiment blog diagram

2
The 2nd International Conference on Engineering and Applied Technology IOP Publishing
IOP Conf. Series: Materials Science and Engineering 674 (2019) 012040 doi:10.1088/1757-899X/674/1/012040

2.4 Frequency to Voltage Converter Circuit Design

In this F/V converter circuit, LM131 IC is used as active component F/V converter output at maximal
input is designed at 10 V, with a maximum of 10% ripple voltage at a speed of 100 rpm. According to
table 1, the maximum frequency is:

20 900 1
f max  hz
rev min 60
f max  300hz
From this maximal frequency, we can derive a minimal time period (Tmin) ) as:
Tmin  1  1  3.3ms
f maks 300
Based on Tmin, we get Tpulse min : 0.5 Tmin = 1,67ms; In the datasheet IC LM131, Touthigh is suggested
at 0.8 from Tmin ([6]Jacob,J.,1989) or 0.8 x3,33 ms = 2,664 ms.

In this study, we use Ct = 0.33µF, so Rt becomes:

touthigh
Rt  (3)
1.1Ct
2,664ms
Rt   7,3K
1.1 x 0.33 F
Rd and Cd determination based on rule 5 Rd Cd<<Tpulsemin (Jacob,J.,1989). Because Tpulsmin is
=1,67mS and Rd= 10kΩ, taken as reference, we can conclude to calculate Cd as :

5R DC D  0.11,67ms  (4)
Using formula 6 above, we get :

CD 
0.11,67ms   03,3nF
510kΩ 
Vave 
2v 1.1R t Ct R Lfin
Rs
V ave is used to determining Rs value. With Vave=10 v, Rs has value (Jacob,J,1989):
(2V)(1.1R t C t )R Lf in
Rs  (5)
Vave
R s  14,8k
As part of F/V converter circuit, a filter is added. This filter is consists of RL CF, filter circuit has
time constant,τ, at τ=RL CF. Vout equation becomes [5] :
In order to get ripple voltage at 10% then :
Vout
 90%  e  τ/T
Vpk
t
ln(0.9)  
R LC
C F  0,72μ,

In this filter circuit, CF=1µF is used.

3
The 2nd International Conference on Engineering and Applied Technology IOP Publishing
IOP Conf. Series: Materials Science and Engineering 674 (2019) 012040 doi:10.1088/1757-899X/674/1/012040

3. Results and Discussion

3.1 Characteristic and Stability test of DC motor simulation and In Proteus

In order to obtain the characteristics of the system, the system is feed with several step voltage. These
step voltages have magnitude at 14V and 15 V. These step voltage response graphs shown in Figure 3
and Figure 4.

Figure 3. Graph Response for 14 V step Figure 4. Graph Response for 15 V step
input input

Figure 4 to 5 visual examination, shows the system is an overdamped system, its means that system
has a dumping ratio greater than 1. Also based on graphics examination, proteus simulation results
give consistent results, the system has a rise time value and setting a relatively constant as shown in
the following table 2.

Table 2. Settling time, tr and ts of DC motor model for various step input
Vstep,V Nsetling ,rpm tr ,s ts ,s
14 940 5.2 5.5
15 1100 5.2 5.5
16 1080 5.2 5.6

3.2 Sampled data from the DC motor model step response.

In order to get step response data sample, motor dc model is feed with 15 V step input and
then data is sampled in 0.25 second. Table 3 shown this sampling result.

Table 3. Data sampled at 0.25 s interval of dc motor model speed when feeding with 15 v step signal
Time, s N, rpm No. N, rpm No N, rpm No N, rpm No N, rpm
0 0 1 6.4900 2 8.8200 3 9.7100 4 9.99
0.25 0.6900 1.25 6.8200 2.25 8.9600 3.25 9.7600 4.25 9.99
0.50 1.5900 1.50 7.1100 2.50 9.0800 3.50 9.7900 4.50 9.99
0.75 2.3400 1.75 7.3600 2.75 9.2000 3.75 9.8100 4.75 10

Data in Table 3 is used for transfer function estimation. Transfer function estimation obtained using
Matlab, through the identification tool application. The model used is a process model for a second-
order system, without delay and overdamped. The transfer function estimation results are:

0.676
Estimated transfer function 
0.583s  2.629s  1
2

where Wn equals to 1, K is 0.676, and Dumping ratio is 1.31.

4
The 2nd International Conference on Engineering and Applied Technology IOP Publishing
IOP Conf. Series: Materials Science and Engineering 674 (2019) 012040 doi:10.1088/1757-899X/674/1/012040

3.3 Transfer function confirmation

Transfer function confirmation test is done by comparing the dc motor response to a step input its
included response from estimated transfer function and response from dc motor model. From both
step response graph, graphically its seen that the value of tr ts and settling consistency between the two
graphics, as we can see in figure 5 and figure 6.

Figure 5. T System response when feeding Figure 6. System response when feeding
with 14v step voltage with 15v step voltage

Table 4 shows the comparison of tr,ts and settling time between estimated transfer function and dc
motor when it is fed with 14V.

Table 4. Comparison table when the system is feed with 14V step voltage
Matlab
Proteus DC
Parameter Transfer Deviation
motor model
function
Nsettling,rpm 9.4 8.6 -8.5%
tr ,s 5.2 5.2 0%
ts ,s 5.5 5.7 3,6%

Based on the value of ts and tr from the three tables above, it can be concluded that the resulting
estimates MatLab function approach with the actual transfer function from a dc motor in proteus
simulation deviation values for tr, , ts and Nsetting quite small at less than 1.5% when it was feed with 15
v input.
Confirmation of the transfer function estimation by transfer functions are also performed by
comparing the actual value of a point per point for the sample time 0.25 seconds when 14 v was used
as input (Table 5).

Table 5. Result when system is feed with 14V step voltage

√(Err2) √(Err2)
Time(s) Simulation Estimation Time(s) Simulation Estimation
(%) (%)
0.25 0.497 0.36 27.57 6 8.59 9.23 7.45
0.5 1.48 1.06 28.38 6.25 8.74 9.34 6.86
0.75 2.02 1.82 9.90 6.5 8.81 9.48 7.60
1 2.96 2.55 13.85 6.75 8.86 9.47 6.88
1.25 3.65 3.23 11.51 7 8.91 9.58 7.52
1.5 4.26 3.86 9.39 7.25 8.99 9.65 7.34
1.75 4.64 4.05 12.72 7.75 9.03 9.71 7.53

5
The 2nd International Conference on Engineering and Applied Technology IOP Publishing
IOP Conf. Series: Materials Science and Engineering 674 (2019) 012040 doi:10.1088/1757-899X/674/1/012040

The average error between real value and estimated value from table 7 above is 8.8%. It means that
estimated transfer function close enough approach real transfer function.

4. Conclusion
Based on the description above, we can conclude that the dc motor model contained in Proteus 7 can
be modeled mathematically using the black-box approach.

References
[1] Ahmed S, Buang, Biao, Shah, Shirish May 2007 Novel Identification Method from step
Response Journal Of Control Engineering Practice Vol.15
[2] Jacob J 1988 Industrial Control Electronics: Application and Design Prentice-Hall New Jersey
[3] Franklin G F, Powell J D, and Workman M L 1990 Digital Control of Dynamic Systems
Addison Wesley 2nd edition
[4] Unbehauen H and Rao G P 1998 A review of identification in continuous-time systems
Annual Reviews in Control vol. 22 p. 145–171.
[5] Ljiung L 1999 System Identification: Theory for the User Prentice Hall, 2nd edition
[6] Ogata K,1990 Modern Control Engineering Prentice Hall Publisher New Jersey p. 57-58
[7] Saab S S and Kaed-Bey R A 2001 Parameter identification of a DC motor: an experimental
approach The 8th IEEE International Conference on vol .2 p. 981-984
[8] Valdivia V, Barrado A, Laazaro, Zumel A, Raga P and Fernandez C 2009 Simple Modeling
and Identification Procedures for “Black-Box” Behavioral Modeling of Power Converters
Based on Transient Response Analysis IEEE Transaction on Power Electronics
[9] Wei Wu 2012 DC Motor Parameter Identification Using Speed Step Responses Modelling and
Simulation in Engineering Hindawi
[10] Labcenter Electronic 2012 Proteus VSM DC Motor Model Manual LabCenter