Designing a Control for a Material Fatigue Testing Machine

TABLE OF CONTENTS

Chapter 1: Project description

Chapter 2: Main concept

Chapter 3: Stepper Motors
3.1 Types
3.1.1 Variable Reluctance Motors
3.1.2 Unipolar Motors
3.1.3 Bipolar Motors
3.1.4 Multiphase Motors
3.2 Stepper Motor Parameters
3.3 Characteristics of the particular motors chosen

Chapter 4: Low-level Control
4.1 H-Bridge
4.1.1 Switching elements: IGBT vs. MOSFET
4.1.2 Low-side Switch
4.1.3 High-side Switch
4.1.4 Circuit design of the H-Bridge
4.2 Short Circuit Avoidance
4.3 Current Control
4.3.1 Pulse Width Modulation
4.3.2 The 555 timer circuit
4.3.3 Reference Voltage

Chapter 5: Mid-level Control
5.1 Moore Machine
5.2 Step Synthesis
5.3 Sequential Logic
5.4 Reset and Enable
5.5 Interface

Intersection: The Low- and Mid-level-PCB

Chapter 6: High-level Control
6.1 Timing Devices
6.1.1 Interrupts
6.1.2 Timing Hardware
6.1.3 Timing Routines
6.1.3.1 The Procedure SetTimerSpeed
6.1.3.2 The Function GrabTimer
6.1.3.3 The Procedure SetTimerHandler
6.1.3.4 The Procedure InternalHandler
6.1.3.5 The Procedure DisableHandler
6.1.3.6 The Procedure NormTimerSpeed
6.1.3.7 The Procedure ResetTimer
6.2 I/O-Management
6.2.1 Developing an ISA I/O-Card
6.2.2 The Parallel Port
6.2.3 I/O -Routines
6.2.3.1 The Procedure MotorWrite
6.2.3.2 The Procedures Motor1Move and Motor2Move
6.3 Control Routines

Chapter 7: Human Machine Interface:
7.1 Turbo Vision
7.2 The TFatigueControl Object
7.2.1 Desktop and Menu Design
7.2.2 Options-Window

Chapter 8: Summary and outlook

Bibliography

Appendix
Complete Source Code
Data Sheets

LIST OF FIGURES

1.1. Biaxial Material Fatigue Testing Machine

2.1. Main Concept

3.1. Schematic of a variable reluctance stepper motor

3.2. Schematic of a unipolar stepper motor

3.3. Schematic of a bipolar stepper motor

3.4. The two control sequences of bipolar motors

3.5. Half step modus

3.6. Schematic of a multiphase stepper motor

4.1. H-Bridge

4.2. Symbol of an n-channel MOSFET

4.3. Symbol of an n-channel IGBT

4.4 H-Bridge circuit diagram

4.5. Forward mode

4.6. Fast decay mode

4.7. Dynamic braking mode

4.8. Short circuit avoidance

4.9. Pulse Width Modulation

4.10. Current mode control

4.11. Pin specification of a 555

4.12. 555 circuit

4.13. Reference voltage circuit

5.1. Moore machine state diagram

5.2. KV-Diagram for X1*

5.3. KV-Diagram for Y1*

5.4. KV-Diagram for X2*

5.5. KV-Diagram for Y2*

5.6. Sequential logic

5.7. KV-Diagram for the decrease signal

5.8. Mid-level control circuits

X.1. Assembly Top Layer

X.2. Top Layer Routes

X.3. Assembly Bottom Layer

X.4. Bottom Layer Routes

X.5. Simulation Result Example

6.1. Interrupt driven program control transfer

6.2. Interrupt Vector Table

6.3. 8253/8254 Programmable Interval Timer

6.4 Circuit diagram for an ISA I/O-card

6.5 Assembly Top View of the ISA I/O-card

6.6 Routes on the top layer of the ISA I/O-card

6.7 Routes on the bottom layer of the ISA I/O-card

6.8 Status window

7.1 Desktop

7.2 Options window

Chapter 1 PROJECT DESCRIPTION

One main field of activity of the Department of Mechanical Engineering and Material Science at the Escuela Superior de Ingenieros de Sevilla is material fatigue testing. In order to evaluate material characteristics under specified conditions, material probes are outset to biaxial movements, i.e.

- traction or pressure, and
- torsion.

The machines for this particular application typically use hydraulics for the power generation; the exact movement is achieved by servo-ventils. Industrial manufacturers demand at least 300.000 € for building a machine like that. This appears very expensive, keeping in mind the very basic task of the machine. Whereas a university of western industrialized countries may bear inversions like that, for universities of third world it is hardly possible to procure such a machine.

This was the motivation for the Department of Mechanical Engineering and Material Science at the Escuela Superior de Ingenieros de Sevilla to start a project in order to find a cheap as well as reliable solution for biaxial material fatigue testing.

The main change in respect to the up to now common machine design (an example is shown in figure 1.1) is that the biaxial movement of the machine to be build shall be realized by two stepper motors. This evokes two main advantages:

- From the mechanical point of view, the power transmission becomes easier: The deviation via the pressure generation is no longer needed, instead the power of the stepper motors can be transformed by screws into the desired movement.
- Regarding the control, stepper motors give the opportunity of a precise movement without the necessity of a control loop.^* Most of all the costs of sensors (most suitable would be LVDT’s) can be avoided in this way.

The project was divided into two main parts, each of them treated by a student with a related field of studies:

- The mechanical part: A student of Mechanical Engineering was to design the mechanical structure of the machine, including the selection of suitable stepper motors.
- The control part: As a student of Electrical Engineering it became my part to design the control for the machine.

The components of the control were not specified. However, certain parts were strongly recommended:

- For high-level control old computers (80286-80486) should be used; these are easily to get for free, as they are no longer useful for modern office applications.
- The control must be simple as well as robust.
- Regarding the low target costs the control should consist of as few components as possible.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 1.1 Biaxial Material Fatigue Testing Machine

Chapter 2
MAIN CONCEPT

In order to reduce the complexity of the given task the control is divided into three categories:

- Low-level control:

It is responsible for the current supply of the motors. Additionally, it includes sequential logic for short circuit avoidance.

- Mid-level control:

It has two main tasks: Keeping track of the motor state and providing four input lines: Clock, Direction, Enable and Reset.

- High-level control:

It gives the reset impulse within an initialisation process and continuously takes three main decisions:

- When does each motor have to receive its next clock impulse?
- In which direction shall each motor move?
- How many steps (or half steps, respectively) are still to be done?

Last but not least, a Human-Machine-Interface is provided. Within a Turbo Vision Desktop the user will be able to give exact movement orders and to be informed about the actual state of the system.

A schematic view of the hierarchic control system is shown in Figure 2.1. The arrows indicate that it is an open-loop control, because there is nothing like a feedback.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 2.1 Main concept

Chapter 3
STEPPER MOTORS

Stepper motors convert electrical pulses received by their excitation windings into discrete angular displacements - commonly called “steps”. In fact, they function as decoders, transforming the digital information received into steps of angular position. Since there is a direct correspondence between the number of input pulses and the number of discrete angular steps taken by the motor, there is no need for a feedback transducer for measuring the angular position of the rotor.¹

Stepper motors have a wide range of applications. Typical examples are numerically controlled machine tool positioning systems, motors of read/write heads of floppy disc drives, paper feed in printers; telescope and antenna controls². Due to their easy control they are preferably used as actuators in automated industrial production processes. The popularity of stepper motors is increasing, because mechanical solutions can be replaced by electronic ones.³ Stepping motors come in a wide range of angular resolution. The coarsest motors typically turn 90 degrees per step, while high resolution permanent magnet motors are commonly able to handle 1.8 or even 0.72 degrees per step.⁴

Technically, stepper motors can be viewed as electric motors without commutators. Typically, all windings in the motor are part of the stator, and the rotor is either a permanent magnet or, in the case of variable reluctance motors, a toothed block of some magnetically soft material. All of the commutations must be handled externally by a motor controller, consisting primarily of a drive logic circuit .⁵ As a series of pulses of voltage is applied to the input of the drive circuit, it feeds suitable currents to the stator windings of the motor to make the axis of the magnetic field step around in synchronism with the input pulses. Depending on the pulse rate and the load torque including that of inertia, the rotor follows the axis of the magnetic field due to the torque produced due to the interaction of the magnetic field caused by the permanent magnet rotor or the variation of reluctance with respect to angular position of the rotor.⁶

3.1 Types

3.1.1 Variable Reluctance Motor s

The rotor is made out of slotted steel laminations and has no winding in it. The stator usually is wound for three phases. The stator windings are excited with the help of an external logic circuit.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 3.1 Schematic of a variable reluctance stepper motor

The cross section shown in Figure 3.1 is of a 30 degree per step variable reluctance motor. The rotor in this motor has 4 teeth and the stator has 6 poles, with each winding wrapped around two opposite poles. With winding

number 1 energized, the rotor teeth marked X are attracted to this winding's poles. If the current through winding 1 is turned off and winding 2 is turned on, the rotor will rotate 30 degrees clockwise so that the poles marked Y line up with the poles marked 2.

Using more motor poles and more rotor teeth allows construction of motors with smaller step angle. Toothed faces on each pole and a correspondingly finely toothed rotor allows for step angles as small as a few degrees.

3.1.2 Unipolar Motors

The Unipolar Stepper motor has 2 coils, simple lengths of wound wire. The coils are identical and are not electrically connected. Each coil has a center tap - a wire coming out from the coil that is midway in length between its two terminals.⁷

Abbildung in dieser Leseprobe nicht enthalten

Fig. 3.2 Schematic of a unipolar stepper motor

The motor cross section shown in Figure 3.2 is of a 30 degree per step permanent magnet or hybrid motor. Motor winding number 1 is distributed between the top and bottom stator pole, while motor winding number 2 is distributed between the left and right motor poles. The rotor is a permanent magnet with 6 poles, 3 south and 3 north, arranged around its circumference.

For higher angular resolutions, the rotor must have proportionally more poles. The 30 degree per step motor in the figure is one of the most common permanent magnet motor designs, although 15 and 7.5 degree per step motors are widely available. Permanent magnet motors with resolutions as good as 1.8 degrees per step are made, and hybrid motors are routinely built with 3.6 and 1.8 degrees per step, with resolutions as fine as 0.72 degrees per step available.⁸

3.1.3 Bipolar Motors

Bipolar permanent magnet and hybrid motors are constructed with exactly the same mechanism as is used on unipolar motors, but the two windings are wired more simply, with no center taps. Thus, the motor itself is simpler but the drive circuitry needed to reverse the polarity of each pair of motor poles is more complex.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 3.3 Schematic of a bipolar stepper motor

The schematic in Figure 3.3 shows how such a motor is wired, while the motor cross section shown here is exactly the same as the cross section shown in Figure 3.2.

An H-bridge allows the polarity of the power applied to each end of each winding to be controlled independently. The control sequences for single stepping such a motor are shown below, using + and - symbols to indicate the polarity of the power applied to each motor terminal:

Abbildung in dieser Leseprobe nicht enthalten

Fig. 3.4 The two control sequences of bipolar motors

While unipolar motors are very popular in the low power segment because of their easy control, bipolar motors are standard for high power applications. They use the current more efficiently, because the current is lead through a whole winding, not only a half as in unipolar configuration. This results in an augmentation of torque by a factor of 2 . Hence for this project there were elected two bipolar motors.

It is possible to get a motor that normally does 200 steps per round to take

0.9° steps, i.e. it will do 400 steps per round. This electronic based technique is called half step modus.⁹

Abbildung in dieser Leseprobe nicht enthalten

Fig. 3.5 Half step modus

3.1.4 Multiphase Motors

A less common class of permanent magnet stepping motor is wired with all windings of the motor in a cyclic series, with one tap between each pair of windings in the cycle. The most common designs in this category use 3-phase and 5-phase wiring. Although these motors can provide more torque from a given package size (because all or all but one of the motor windings are energized at every point in the drive cycle) and some 5-phase motors have high resolutions on the order of 0.72 degrees per step, they are not very common in use. The reason can certainly be found in the more difficult control, which requires 1/2 of an H-bridge for each motor terminals.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 3.6 Schematic of a multiphase stepper motor

3.2 Stepper Motor Parameters

The most important parameters of stepper motors are as follows:¹⁰

- Step angle, defined as the angular displacement of the rotor in response to each pulse;
- Holding torque representing the maximum load torque at which the rotor does not move when an exciting winding is energized;
- Limiting Torque, defined for a given pulsing rate or stepping rate measured in pulses per second, as the maximum load torque at which the motor follows the control pulses without missing any step;
- Synchronous stepping rate defined as the maximum rate at which the motor can step without missing steps. The motor can start, stop or reverse at this rate.
- Slewing rate, defined as the maximum rate at which the motor can step unidirectionally. The slewing rate is much higher than the synchronous stepping rate; but the motor will not be able to stop or reverse without missing steps at this rate.

3.3 Characteristics of the particular motors chosen

The selected stepper motors’ specification is as follows :

Abbildung in dieser Leseprobe nicht enthalten

Remark:

Principally, there is a choice between using servomotors and stepper motors, as both types of motors offer similar opportunities for precise positioning. However, they differ in very important fact: Servomotors need closed loop control, which requires analogue feedback control systems of some type. Stepper motors can be used in simple open-loop control systems. If a stepper motor in an open-loop control system is overtorqued, all knowledge of rotor position is lost and the system must be reinitialised; servomotors are not subject to this problem.

However, already at the very beginning of the project the use of stepper motors was a datum.

Chapter 4
LOW-LEVEL CONTROL

4.1 H-Bridge

As bipolar stepper motors have no center taps on their windings, the current through the winding needs to be reversed, in order to reverse the direction of the field produced by a motor winding. This is realized in form of an H bridge¹¹ as shown in Figure 4.¹¹²:

Abbildung in dieser Leseprobe nicht enthalten

Fig. 4.1 H-Bridge

H-Bridges for smaller stepper motors (up to approx. 50V and 4A) are already available as integrated circuit solutions. However, the above selected motors exceed by far the characteristics of such an IC, so there remains only the possibility of a discrete design. Certainly the switching elements are the most important components, so the next subsection deals with them.

4.1.1 Switching elements: IGBT vs. MOSFET

Basically, there are two different discrete elements that can be used to re alize a switch for 8A at 160V: Power Metal Oxide Semiconductor Field-Effect Transistors (Power MOSFET) and Isolated Gate Bipolar Transistors (IGBT).

Abbildung in dieser Leseprobe nicht enthalten

Fig. 4.2 Symbol of an n-channel MOSFET

The MOSFET is a normally off device in which the current is transported by majority carriers. A power MOSFET is fabricated from several parallel connected MOSFET cells organized in various geometrical designs. MOSFET cells contain an integral pn-n+ diode between the drain and the source terminal, which is called body diode (or parasitic diode). It allows reverse current to flow when the MOSFET is in the off-state.

Abbildung in dieser Leseprobe nicht enthalten

Fig. 4.3 Symbol of an n-channel IGBT

The insulated gate bipolar transistor, as the name suggests, is a bipolar transistor with an insulated base terminal (º gate). The insulated gate enables simpler gate drive circuits like those in MOSFET. An IGBT does not conduct current in the reverse blocking mode when a negative voltage is applied to the collector with respect to the emitter.

In the power MOSFET, the decrease in the electron mobility with increasing temperature results in a rapid increase in the on-state resistance of the channel and hence the on-state voltage drop. In IGBT this increase is very small. The on-state voltage drop in power MOSFET increases by a factor of 3 between the room temperature and 200°C. Whereas in the identical conditions in the IGBT, the increment in the on-state voltage drop is very small.

The turn-on transients are identical in MOSFET and IGBT. In the turn-off transition, however, the IGBT has a large current tail due to slow removal of the charge carriers.¹³

It can be stated that the power MOSFET is suited for applications that require low blocking voltages and high operating frequencies. IGBT is the preferred device for applications that require high blocking voltages and lower operating frequencies.

In the final circuit of the H-Bridge MOSFETs as well as IGBTs are used (see Fig. 4.3).

Remark: As the H-Bridge is completely symmetric, in the following two subsections only the left side of the H-Bridge will be explained.

4.1.2 Low-side Switch

The realization of the low-side switches is very simple. The logic input signal at B is directly lead to the gate of an n-channel MOSFET. The BSS123 has been chosen for this particular task, because its gate is TTL compatible. This MOSFET together with the voltage divider built by R9 and R11 aliments the gate of an n-channel IGBT with either 12 V or 1.1 V. As IGBT an IRGBC30 has been selected, because its rated current as well as its backward voltage satisfy the needs.

There are two main reasons for the use of an IGBT in this case:

- The operating frequency for the low-side switch equals the motor’s stepping frequency, which will hardly be more than 5 kHz. Under these circumstances the IGBT’s advantage of low static power dissipation outpaces by far the MOSFET’s advantage of low switching losses.
- If the switching transistor (either the IGBT or MOSFET) and the power diode were within one single component (that is the case with the parasitic diode in a MOSFET), it would be impossible to have a sense resistor placed like in the actual circuit (figure 4.3). As a consequence, the current through the motor winding would not always circulate through the sense resistor, but only during the switch-on period!

As a result of the actual design of the low-side switch, B is inverted. I.e., when the logic input signal is high, the related IGBT will not conduce and the switch is open!

4.1.3 High-side Switch

Designing a circuit for the high-side switch is by far more sophisticated, because the voltage potential of the main switching transistor’s source (or emitter, if it was an IGBT) is alternating between GND and +160 V, depending on the actual state of the motor driver. It is indispensable to provide a control voltage of at least 10 V between gate and source for switching on the transistor.

In the present H-Bridge a so called bootstrap circuit has been realized. Proceeding from the situation that M1 once has been switched on, the point

W will almost have GND potential. The capacitor C1 is being charged via D1 up to approx. 11 V. When M1 is switched off and A becomes high, a current will flow through R7, Q5 and D3 into the basis of Q1. Q1 is turned on a nd puts C1’s voltage at the gate of M1. M1 is switched on, and the point W is elevated up to +160 V. C1 is “taken along”, supplying the driver circuit during M1’s switch-on period. Q5 can be regarded as current source and keeps on the control of M1. In order to open the switch again, a low-signal is given to A. The current through Q5 is turned off, withdrawing the basis current from Q1. Due to the anti-saturation wiring built by D3 and D5 the transistor Q1 is switched of without any delay worth mentioning. Q3 now shorts out the voltage at the gate of M1 and thus turning it off. The current still circulating through the motor winding gives rise to a near GND potential at the point W. The capacitor C1 is being charged via D1 up to its full capacity again, thus being prepared for switching on M1 for the next time.

As the capacitor C1 is slowly being discharged by the driver circuit, M1 cannot keep an on-state for too long a time. In order to avoid a static operation of M1, the high-side switch receives the task of current control by pulse width modulation (see also 4.3 and especially 4.3.1).

The motivation for using MOSFETs stands in exact contrary to the argumentation in the subsection before:

- Being switched by the current control (pulse width modulation), the high-side switch operates above audible frequencies (>20kHz), thus avoiding static operation. Under these circumstances the low switching losses of the MOSFET are decisive.
- As the MOSFET already includes a parasitic diode, no additional diode is necessary, thus reducing parts count and cost.

4.1.4 Circuit design of the H-Bridge

Figure 4.4 shows the final realization of the H-Bridge. The motor winding is approximated with an inductor (LL1) and a resistor (RL1). RS1 is the sense resistor, which allows to measure the actual current through the winding. The four connections A, B, C and D are the TTL-compatible inputs to control the switches.

The symmetric design of the H-Bridge is clearly visible. It has to be kept in mind that for each motor two H-Bridges are needed (one for each winding).

Abbildung in dieser Leseprobe nicht enthalten

Fig. 4.3 H-Bridge circuit diagram

[...]

---

^* Chapter 3 will explain the principle of stepper motors and their advantages.

¹ Pillai, S. K. A First Course on Electrical Drives, 2nd ed. New Delhi et al.: Wiley Eastern, 1989, p. 193

² Ramírez, Javier. Controlador de motor paso a paso. In: electrónica, 2nd vol. 2001, p. 40

³ Kreuth, Hans-Peter. Elektrische Schrittmotoren. Sindelfingen: expert-Verlag, 1985, p. 41

⁴ Jones, Douglas W. Control of Stepping Motors: a tutorial, 2nd ed. IOWA City, IA: 1998, http://www.cs.uiowa.edu/~jones/step/types.html

⁵ Jones, Douglas W. Control of Stepping Motors: a tutorial, 2nd ed. IOWA City, IA: 1998, http://www.cs.uiowa.edu/~jones/step/index.html

⁶ Pillai, S. K. A First Course on Electrical Drives, 2nd ed. New Delhi et al.: Wiley Eastern, 1989, p. 193

⁷ Laidman, R. Unipolar Stepper Motors and Control, http://206.96.14.115/stepper/Tutorials/UniTutor.htm 8

⁸ Jones, Douglas W. Control of Stepping Motors: a tutorial, 2nd ed. IOWA City, IA: 1998, http://www.cs.uiowa.edu/~jones/step/index.html

⁹ Maloney, Timothy J. Modern industrial electronics, 4 t h ed. Upper Saddle River, NJ: Prentice-Hall, 2001, p.575

¹⁰ Pillai, S. K. A First Course on Electrical Drives, 2nd ed. New Delhi et al.: Wiley Eastern, 1989, p. 199

¹¹ Not only are H-bridges applied to the control of bipolar stepper motors, but also to the control of DC motors, push-pull solenoids (those with permanent magnet plungers) and many other applications.

¹² Jones, Douglas W. Control of Stepping Motors: a tutorial, 2nd ed. IOWA City, IA: 1998, http://www.cs.uiowa.edu/~jones/step/circuits.html

¹³ Agrawal, Jai P. Power electronic systems. Theory and design. Upper Saddle River, NJ: Prentice-Hall, 2001, 134 16