Design of A Microprocessor-Based Control System of A Compression Molding Process George K. Adam
Design of A Microprocessor-Based Control System of A Compression Molding Process George K. Adam
net/publication/2921274
CITATIONS READS
0 3,804
1 author:
George Adam
University of Thessaly
93 PUBLICATIONS 247 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by George Adam on 24 September 2015.
George K. Adam
Technological Educational Institute of Larissa, Larissa, Greece
gadam@teilar.gr
Proceedings of the Ninth IEEE International Conference on Engineering Complex Computer Systems Navigating Complexity in the e-Engineering Age
1050-4729/04 $20.00 © 2004 IEEE
system is driven by a large hydraulic pump that requires storage of the control data. A simplified block diagram of
240Vac (rated at 20A). the 80C188EB-based controller is given in Figure 2.
High-level user interface routines under develoment Mold unit actuator Bridges
mold valve switches
M
pos relay relays
Software
Serial communication Compression control developed
assembly subroutines assembly subroutines
24VDC I/O sockets
Hardware
purchased 82C55 PPI ADC & DAC RS-232
80C188EB-based Process Controller
or built
Static RAM
2x32KByte
EPROM
64KByte
Intel
Press Hydraulic 80C188EB
other
machine equipment
CPU
Hardware and equipment supplied by the manufacturer
Software interface and application program
Figure 1. Developed and proprietary hardware Port 82C55 Parallel I/O Ports
and software. C A&B
Keyboard Indicators
0 1 2 3 4
The control of the compression molding process is 5 6 7 8 9
A B C D E
based on sensory information which is obtained from
mold’s position sensors. In this way, data are collected Figure 2. Block diagram of the 80C188-based
from the process and used as input to the software control controller.
algorithm. Sensors allow the core control unit (80C188EB
processor) to detect the state of the compression molding 3. Controller design architecture
process (e.g., the inductive proximity sensors sense
whether a metal object passes nearby). When a sensor The core of the control part is the 80C188EB
detects a change it signals that change to the controller. microprocessor the architecture of which was optimized
This is typically done by switching a voltage or current on for applications in industrial controllers. The use of the
or off. The controller processes the inputs of sensors microprocessor reduces the amount of elements required,
feedback and controls the electrical valves that drive the simplifying the design process, and subsequently the
equipment (hydraulic pump motor and compression board's dimensions.
actuators). Typical process parameters that are inputs to The initial design development procedures of the
the controller are mold position (based on mold limit controller and control software were followed by
switch), valve status (opened or closed), and motor on or simulations and real-time implementation. The
off. The valves control the rate of the hydraulic fluid flow, development of the controller’s physical layout design
through the hydraulic solenoids to the actuators that was implemented using OrCAD Capture software
perform mold compression and product extraction. A (Cadence Design Systems, Inc.).
compression mold operation here produces multiple parts
(concrete products) on each work cycle. 3.1. The control hardware
Proceedings of the Ninth IEEE International Conference on Engineering Complex Computer Systems Navigating Complexity in the e-Engineering Age
1050-4729/04 $20.00 © 2004 IEEE
32MHz into three ports (A, B, C) of 8 lines each. This circuit is
80C188EB software programmable for three modes of operation.
28 61
27 P1.0/GCS0 AD0 66
200p 20p 20p P1.1/GCS1 AD1
26 68
25 P1.2/GCS2 AD2 70
4.5u 24 P1.3/GCS3 AD3 72
21 P1.4/GCS4 AD4 74 A0-A15
20 P1.5/GCS5 AD5 76
19 P1.6/GCS6 AD6 78
P1.7/GCS7 AD7 62
58 A8 67 10 11 10 11 10 11
P2.1/TXD1 A9 2 19 9 A0 O1 12 9 A0 D0 12 9 A0 D0 12
55 69 D0 Q0 A1 O2 A1 D1 A1 D1
64Kx8 EPROM
32Kx8 SRAM
32Kx8 SRAM
P2.3/SINT1 A10 3 18 8 13 8 13 8 13
71 D1 Q1 A2 O3 A2 D2 A2 D2
Address Latch
A11 4 17 7 15 7 15 7 15
57 73 5 D2 Q2 16 6 A3 O4 16 6 A3 D3 16 6 A3 D3 16
P2.0/RXD1 A12 D3 Q3 A4 O5 A4 D4 A4 D4
A0-A7
50 75 6 15 5 17 5 17 5 17
P2.6 A13 7 D4 Q4 14 4 A5 O6 18 4 A5 D5 18 4 A5 D5 18
49 77 D5 Q5 A6 O7 A6 D6 A6 D6
P2.7 A14 79 8 13 3 19 3 19 3 19
9 D6 Q6 12 25 A7 O8 25 A7 D7 25 A7 D7
52 A15 80 D7 Q7 A8 A8 A8
TXD0 A16 24 24 24
81 11 21 A9 21 A9 21 A9
4 3 44 A17 82 1 LE 23 A10 23 A10 23 A10
19 R1IN R1OUT 20 CLKOUT A18 83
OE 2 A11 2 A11 2 A11
R2IN R2OUT A19/ONCE 26 A12 26 A12 26 A12
10 74HC573/LCC A13 A13 A13
2 5 9 S0 53
27
A14
1
A14
1
A14
T1IN T1OUT S1 RXD0 1
1 18 8 2 18 A15 22 22
T2IN T2OUT S2 36 3 A0 B0 17 22 27 OE 27 OE
8 6 PDTMR 4 A1 B1 16 20 OE/VPP 20 WE 20 WE
Data tranceiver
C1+ ALE 5 A2 B2 15 CE VCC CE VCC CE
13 11 41 A3 B3
D0-D7
11 C1- 16 DEN CLKIN 40
6
A4 B4
14 TMS27C512 28
VCC
28
VCC
C2+ DT/R OSCOUT 7 13
15 12 8 A5 B5 12
10 C2+ 15 HLDA 31 9 A6 B6 11
HM62256 HM62256
16 C2- 4 LOCK INT0 32
A7 B7
C2- RD INT1 1
D0-D7
14 38 35 DIR
12 V+ 7 RESOUT INT4 19
OE
17 V- 45 RFSH 59
V- 47 T0O P2.2/BCLK1 56
74HC245
5 T1O P2.4/CTS1 54
MAX233 30 WR P2.5/BCLK0 VCC 2 18
A1 Y1
29 UCS 51 3 17
4 A2 Y2 16
LCS CTS0 13 A3 Y3
5 15
serial 33 HOLD 17 6 A4 Y4 14
1 34 INT2/INTA0 NMI 18 7 A5 Y5 13
INT3/INTA1 RDY 8 A6 Y6 12
6 37 A7 Y7
2 RESIN 46 9 11
A8 Y8
7 T0I 48
T1I 1
3 14 19 G1
8 TEST G2
4 74HC541
9
5
Figure 4. Memory units.
CONNECTOR DB9
Proceedings of the Ninth IEEE International Conference on Engineering Complex Computer Systems Navigating Complexity in the e-Engineering Age
1050-4729/04 $20.00 © 2004 IEEE
The analog-to-digital converter (ADC0804) converts
ADC0804 24VDC Input Socket O1 to hydraulic pump
18 6 2 1 relay output
17 DB0 +IN 7 4 3 3 1
16 DB1 -IN 6 5 2
15 DB2 9
14 DB3 VREF/2
PIA 82C55 13 DB4 4
solid state mold sensor
34 4 12 DB5 CLKIN 19
33 D0 PA0 3 11 DB6 CLKR 1K
32 D1 PA1 2 DB7 2
31 D2 PA2 1 5 RD 3
0.001u
30 D3 PA3 40 INTR WR 1
29 D4 PA4 39 CS
28 D5 PA5 38 20
27 D6 PA6 37 VCC/VREF
D7 PA7
9 18
8 A0 PB0 19
A1 PB1 20
35 PB2 21 7 11
24VDC Output Socket
5 RESET PB3 22 6 DI0 IOUT1 12 2 1
to the solenoid valve
36 RD PB4 23 5 DI1 IOUT2 4 3
O2 (drives the compression)
6 WR PB5 24 4 DI2 9 6 5
CS PB6 25 16 DI3 RFB
PB7 15 DI4 24V DC ac motor
14 14 DI5 K1
PC0 15 13 DI6 1 2
PC1 16 VCC DI7
PC2 17 1
PC3 13 2 CS
PC4 12 18 WR1 240Vac
PC5 11 19 WR2 Motor relay SPST
PC6 10 17 ILE power supply
PC7 XFER
8
VREF
DAC0830
2 1
O3
4 3 other
O4
6 5
O5 sensor relays
3 4 3 4 3 4 3 4 3 4 3 4
interrupt signal (INTR) here is not shown it is being 5 6 5 6 5 6 5 6 5 6 5 6
connected to an interrupt input of the 80C188EB 7 8 7 8 7 8 7 8 7 8 7 8
conversion. The relay output of the hydraulic pump is 11 1211 1211 1211 1211 1211 12
2
7-segment LED displays (LN07402) interfaced (via ULN 7406
2003A) to port A, and a small keyboard of 16 switches (0-
1
9, A-F) interfaced to ports B (PB0-PB1) and C (PC0-PC5) PB2 PB3 PB4 PB5 PB6 PB7
of the 82C55. The controller and control software Figure 6. Indicators display.
generates periodical time interrupts to determine whether
any of the keys was pressed. This is initiated by an 3.2. The control software
interrupt generated from the internal timer of the
microprocessor every 600µs. In this way during a single The control software was written in assembly language
interrupt the keyboard is checked and the indicators are using the 80x86 family MASM Assembler. The code was
updated, while at the same time the sensors feedback is optimized in order to use efficiently the available memory
checked and the hydraulic pump’s (valves) operation is and reduce compilation time and execution speed. The
regulated. Figure 7 shows the design of the keyboard program is designed to respond to sensor or operator input
interface. by executing the appropriate subroutine corresponding to
The voltage levels for the inputs and outputs are the input. This program of instructions is repeated on each
important. Although the digital outputs (TTL level) will work cycle. Each work cycle consists of the same steps
be between 0V and 5V, analog inputs and outputs will and associated process parameters. An indicative
vary. For this reason if these voltage limits are exceeded fragment of this control software module is given below.
the board has built in protection (1A fast blow fuse). Here, the controller compares the sensor output (mold
Proceedings of the Ninth IEEE International Conference on Engineering Complex Computer Systems Navigating Complexity in the e-Engineering Age
1050-4729/04 $20.00 © 2004 IEEE
PIA
34 4 1 16
40
33 D0 PA0 3 2 1B 1C 15
32 D1 PA1 2 3 2B 2C 14
31 D2 PA2 1 4 3B 3C 13
30 D3 PA3 40 5 4B 4C 12
29 D4 PA4 39 6 5B 5C 11
28 D5 PA5 38 7 6B 6C 10
27 D6 PA6 37 7B 7C
D7 PA7 9
2 19 9 18 COM
3 D0 Q0 18 8 A0 PB0 19
D1 Q1 A1 PB1 ULN2003A
Address Latch
4 17 20
5 D2 Q2 16 35 PB2 21
D3 Q3 RESET PB3
A16-A17
6 15 5 22
7 D4 Q4 14 36 RD PB4 23
8 D5 Q5 13 6 WR PB5 24
9 D6 Q6 12 CS PB6 25
D7 Q7 PB7
11 14
PB0-PB7
1 LE PC0 15
OE PC1 16
PC0-PC5
PC2 17
PB0 PB1 VCC +5V
74HC573/LCC PC3 13
PC4 PC5
PC4 12 1
7 2 1
6 2 1
F 2 1
E 2
PC5 11
PC3 10K
PC6 10
PC7 1
5 2 1
4 2 1
D 2 1
C 2
PC2
82C55
1
3 2 1
2 2 1
B 2 1
A 2
PC1
1
0 2 1
1 2 1
8 2 1
9 2
PC0
Proceedings of the Ninth IEEE International Conference on Engineering Complex Computer Systems Navigating Complexity in the e-Engineering Age
1050-4729/04 $20.00 © 2004 IEEE
The verification process took about 40% of the design decreased. In addition, the system employs only those
cycle. The verification test generates input signals and circuits that are needed for the implementation of process
checks the value of the output signals. A fragment of the monitoring and control. An accurate analysis and
simulated test data used is the following: evaluation of the controller was performed by using
Inputs: D0-D7: bit_vector(7 down to 0) := “11111111” simulation software. Although several problems have
Outputs: PB0-PB7: std_logic_vector((7 down to 0) been tackled, the results obtained of the implementation
The input vector PA0-7 reads the same value if the test is of the presented control system were satisfactory, and in
passed. some cases improved the operation accuracy of the system
under control. In addition, certain key points for further
5. Performance results research have been identified.
In any automated system may occur malfunctioning
The performance of the designed control system was events that may result in costly delays and loss of
evaluated under real-time operating conditions. Real-time production. For this reason, among other objectives (e.g.,
experiments were conducted to observe the functionality improve high-level control, based on previous research
of the controller during the operation of the compression work [8]), future work will focus in improving the self-
molding system. All the tests and measurements have diagnostic functions, such as monitoring for error
been carried out based on constant compression pressure, detection and recovery, in order for the controller to
as well as other specific manufacturing factors and automatically take the necessary corrective actions.
conditions. During the real test procedures several system
internal data were collected and analyzed for testing 7. References
primarily system's efficiency and accuracy. Although real-
time experiments were time-consuming, certain tests have [1] J. Henkel, “Closing the SoC Design Gap”, IEEE Computer,
been performed and the results obtained were compared IEEE Computer Society Press, Los Alamitos, CA, September
to the specifications established during the initial design 2003, pp. 119-121.
stages. Although some of the initial design specifications [2] Bose, B.K., Microcomputer Control of Power Electronics
and Drives, IEEE Press, New York, 1987.
had to be redefined and redesigned, the final control [3] Bollinger, J.G., and N.A. Duffie, Computer Control of
system proved to be stable, ensuring normal operation of Machines and Processes, Addison Wesley, New York, 1988.
the process under control. [4] Astrom, K.J., M. Blanke, A. Isidori, W. Shaufelberger, and
R. Sanz (eds.), Control in Complex Systems, Springer Verlag,
6. Conclusion New York, 2000.
[5] Holt, D.R., Integrated Manufacturing Engineering Systems,
The design of complex control systems requires McGraw-Hill Inc., New York, 1992.
[6] G.N. Saridis, “Intelligent Manufacturing in Industrial
efficient and powerful CAD tools, which if applied
Automation”, Handbook of Industrial Automation, R. Shell, and
correctly may have a considerable positive influence on E. Hall, (eds), Marcel Dekker, New York, 2000, pp. 485-488.
quality, development time and cost of the embedded [7] Short, K.L., Embedded Microprocessor Systems Design: An
product. Introduction Using the Intel 80C188EB, Prentice-Hall, New
In this paper was presented the design of a low-cost York, 1998.
efficient controller of a compression molding system. The [8] G.K. Adam, and N. Mastorakis, “High level programming
controller was realized on the base of the 80C188EB Intel and control of a manufacturing system”, Transactions on
embedded microprocessor. It is evident, that by using on- Circuits, vol. 2, no. 1, WSEAS, New Jersey, January 2003, pp.
board a microprocessor the amount of electronic elements 301-308.
implemented in the design and final realization is
Proceedings of the Ninth IEEE International Conference on Engineering Complex Computer Systems Navigating Complexity in the e-Engineering Age
1050-4729/04 $20.00 © 2004 IEEE