0% found this document useful (0 votes)
6 views22 pagesRobotics Module 3- Drives and Actuators
Actuators are devices that provide motive force for robotic joints, classified into pneumatic, hydraulic, and electric types based on their power input. Hydraulic systems offer the highest power-to-weight ratio and are known for their stiffness and reliability, while pneumatic systems are lighter and more compliant. The document also details the characteristics, classifications, and performance of hydraulic and pneumatic actuators, including design considerations and advantages of various types.
Uploaded by
anbattulvenugopalCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
6 views22 pagesRobotics Module 3- Drives and Actuators
Actuators are devices that provide motive force for robotic joints, classified into pneumatic, hydraulic, and electric types based on their power input. Hydraulic systems offer the highest power-to-weight ratio and are known for their stiffness and reliability, while pneumatic systems are lighter and more compliant. The document also details the characteristics, classifications, and performance of hydraulic and pneumatic actuators, including design considerations and advantages of various types.
Uploaded by
anbattulvenugopalCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 22
Actuators are the devices which give the actual mative force
for the robotic joints. Generally, power is supplied by
Compressed air, pressunzed fluid or electricity
Depending upon the type of power input to actuator, they are
alled a5 pneumatic actuators with compressed air as input
Power, hydraulic actuators with pressurized fluid as input
ower and electric actuators with electric input
.2_ CLASSIFICATION OF DRIVES
Cae ay
ot
a
oJ fe
tinea Rory oy
sconces aay open, von
ctnygnie otinae
4
ct
‘sage Dobie Gow Vine Rae
wetng sig mone nour end pron
‘The drives are chosen based on =
(The power consumption
(i) Positional accuracy
Git) Repeatability
GW) Speed of operation
() Stabitiry
(i) Reliability
ost
j-3_ CHARACTERISTICS OF DRIVES:
() Welght: It is an important characteristics of any actuating
system, It plays important role while selecting actuating
system for a particular application,
(I) Power to Weight Ratio: It shows the type of actuation
system which will be best suited for given application and
higher the power to weight ratio, better is suited for given
application.
+ Power to weight ratio of electric systems is average.
© Stepper motors are generally heavier than
sevomotors for the sume power and thus hive 3
lower power to weight rati.
© Pneumatic cyfinders deliver the lowest power to
weight ratio
CHAPTER 4
DRIVES AND CONTROL SYSTEMS
* Hydraulic systems have the highest power to weight
ratio,
Gi Operating Pressure: Hydraulic system delivers high
Power, dus to high operating pressures ranging from 55
Psi to S000 psi whereas pneumatic system has low
‘operating pressure as compared to hydraulic system that
‘may range between 100 t0 120 psi
OW) Stiffness: It is the resistance offered by a material against
eformation, Stiness és directly related to the modulus of
elasticity of the material. Modulus of elasticity of fluids is
very high, therefore, hydraulic systems are wery stiff, Stitf
systems have a rapid response to changing loads and
Pressures and thewe are more accurate also. But stif
system will not prevent damage to the robot or the part.
(1 Compliance: A system Is said to be compliant, If it
deforms easily under changing load or changing driving
force. Thus, the more compliant system, the easier it
Geforms under the load.
Therelore, hydraulic systems are non-compliant and the
Pneumatic systems are compliant Also a compliant
system will be inaccurate, but it will prevent damage to
the robot or the part
‘Therefore, good working balance is nesced between
sitress and the compliance,
(w Reliability: Hydraulic syctems produce very large forces
with short strokes, thus they can be directly attached to
the links and do not require reduction gear trains. This
Simplifies the design, reduces the weight and thus
increases reliability of the system.
(utl)Resokution: The electric motors rotate at high speeds and
hey must be used in conjunction with the reduction
(gears to increase the torque and to decrease their speed.
‘This increases the cost, and also the number of parts, thus
resus in decrease of reliability and increase the
Fesclution of the system, as itis possible to rotate the link
at very small angle
(vil) Uses of Reduction Gears:
© Hydraulic systems da not need reduction gears for
the power transmission.
+ Bectric systems use reduction gears which réduce the
inertia on the motor.
a
© scanned with OKEN Scanner RONOTICS (BE MECHANICAL)
441 Hydraulic Actuators
Hydraulic actuators are powered by pressurized of fie. at
£1,000 0 3,000 tovin’).
‘They extract energy from the fuid and convert it to
mechanical energy to petorm useful wort,
Depending upon the mation they transmit, the hydraulic
actuators are classified as -
o
Unear Actuators or Hydraulle Cylinders
They convert hydraulic energy into straight line motion
‘hus actuating a linear joint by means of # moving piston,
Simplest form of hydraulic cylinder is the single acting
design, a8 shown in Fig. 4.1.
I consists of a piston inside a eylincrical barre, Attached
tO one end of the piston is a rod, which extends outside
Cone end of the cylinder. At the other end is 2 part for the
entrance and enit of cil. These cyincers can exert a force
in only the extending directon, Retraction is
accomplished by using gravity or by the inclusion of
compression spring in the rod end.
Fig. 42: Double acting cylinder with single piston
In this arrangement, the fluid may enter through ether
the blank end (L.A) or the rod end (Le 8). Extension of
the piston rod occurs when the fluid is pumped into the
blind end. Retraction of the piston occurs when the fusid
is pumped into the rod end of the cylinder.
DRIVES AND CONTROL SYSTEMS
Due to differential area, the retraction stroke is faster than
the extension strove as well force obtained in extension
stroke is more than that of in retraction stroke.
Double Acting with Double Piston Rod:
®)
A Sma 8
Fig. 43: Double acting cylinder with double piston rod
* The piston rod i extended due to the fluid pressure
exposed on the blind end of the piston.
The area exposed 10 the Mluid towards the rod end (ie. B)
of the cylinder reduces due to the presence of the piston
fod and so the ertending force is greater than the force af
retraction.
But the speed of retraction is larger as compared to the
Speed of extension,
* Force and speed are equal in both the directions due to
the extension of piston rods on both sides of the piston
inside the cylinder.
Design Considerations for Hydraulic Cylinder:
Following are the design considerations for hydraulic lind.
(a) Thrust required
(©) Operating pressure
(©) Speed
() Stoke length
(©) Type of construction
(0) Type af mounting
(g) Expected service life
(M) Operating environment
(Lateral loads and buetding failure
(i) Hydraulic Rotary Actuators or Hydraulic Motors
# Rotor actuator is a device that converts hydraulic energy
into rotational motion, it may be limited rotation or
continuous full rotation.
The continuous full rotation type of rotary actuator is
sometimes called as hydraulic motor.
Classification of Hydraulic Motors
Rotary actuators are mainly of three types:
(2) Gear motors
2) Vane motors and
@) Piston motors
© scanned with OKEN Scanner ROROTICS (BE MECHANICAL)
Q) Gear Motors:
These devices work opposite to that of gear pumps and
develop torque and rotary motion when they are acted
upon by the fui
© Fluid enters the inlet pom and is camied around the
outside of the casing and finally flows out of the outlet
port
The direction of rotation of gears can be reversed by
changing the direction of inlet and cutiet flows.
‘+The volumetric displacement of a gear motor is fixed. Side
thrust occurs cue to difference in pressure between inlet
and outlet ports
‘©The shaft drives the load, when it gets power trom one of
the gears in the gear motor.
Fig. dt: Gear motor
Gear motor develops torque due to hydraulic pressure
acting on the surfaces of the gear teeth.
Gear motors are normally limited to 2000 psi operating
Pressures and 2400 rpm operating speeds.
© Gear motors are available with a maximum flow capacity
of 150 gpm.
Advantages:
= Iris simple in design.
# Cost of gearmotor is tow.
+ Ih case of intemal gearing arrangement instaady of
extemal gearing, it can operate at higher pressures and
speeds and also has greater displacements than the
extemal gear motor.
(2) Vane Motor:
Principle: "Vane motors develop torque by the bycraufic
pressure acting on the exposed surfaces of the vanes, which
slide in and out of the rotor connected to the drive shaft
Fig. 45 shows constuctional details af wane motor
un
DRIVES AND CONTROL SYSTEMS
ig. 45:4 Vane motor
ta) Vane subjected to high presture at inlet side and low
pressure opposite
(b) The inlet connecting two opposing pressure passages to
balance side loads on the rotor,
AAs rotor revolves, the vanes follow the surface of the cam
fing because springs are used to force the vanes radially
outward.
There is no centrifugal force until the rotor starts 10
revolve
The sliding action of the vanes forms sealed chambers,
which camry the fluid from the inlet to the outlet.
Balanced design of vane motors is usually followed in
pracuse. In this design, pressure build up at either port is
Girected to two interconnected cavities located 180"
apa.
The side loads that are created are therefore cancelled
UL AS the vane motors are hydraulically balanced, they
are fixed displacement units.
Advantages:
Compared to gear motors, vane motors have less leakage
tendency
Vane motors can be used at lower speeds due to tess
leakage tendency
[AS Side loading occurs on the shaft of a single va
‘motor, these are commonly used.
Je displacement
Piston motors are classified as —
© Inline piston motor
(Radial piston motor.
In-Line Piston Motor:
They generate torque by pressure acting on the ends of
pistons reciprocating inside a cylinder block.
Fig, 4.6 shows the in-fine piston motor in whieh the motor
drive shaft and cylinder block are centered on the same
ais
© scanned with OKEN Scanner DRIVES AND CONTROL SYSTEMS
ROBOTICS (DE MECHANICAL)
ee
bere ee
[ , ==.
: EEE,
ae y |
!
/
om Vi Come ee
, sae pte
1 aac ewes piate angie
Fp 46: Piston motor
Pressure acting on the ends of the pistons generates a
force against an angled swash plate, this causes the
finder block to rotate with a torque that is proportional
to the area of the pistons,
The torque capability is also a function of the swash plate
angle.
le T= 1
‘The swash plate angle gives the volumetric displacement.
Radial Piston Motor:
Radial piston motor is most commoniy used motor. It
operates in reverse of the radial piston pump.
In this type, fluid is forced into the cyfinders and crives
the piston ourmard.
Fig, 4.7 shows the constructional cetais and operation of
radial piston motor:
Fig. 4.7: Radial piston motor
‘The piston pushing against the rotor causes the cylinder
block to rotate. Usually, the motor contains seven or nine
pistons, but for simplicity we have considered three
pistons.
When fluid is forced into the eyfinder bore containing
piston (a), the piston moves outward, 25 the fuid cannot
be compressed. This causes the cylinder to rotate in a
clockwise direction.
[As the force acting on piston (a) causes the cylinder block
to rotate, piston (b) starts to rotate and approach the
Piston of piston (cb
# As piston (b) rotates, iti forced inward and thus, forces
the fad out of the cylinder.
‘© AS there is litte or no pressure on this side of the valve,
the piston is moved easly in by its contact with the
reacton ring of the rotor, the fluid is easly forced out of
the cylinder block and back to the reservoir.
* The action continues as long as fluid under pressure
enters the cylinders.
Advantages of Hydraulle Motors
© Compact In Stre: It has highest power to weight ratio,
it compact in size.
© Wide Range of Speed: i has speed range from zero to
maximum, which can be obtained from the maximum
flow provided by pumping system,
© Frequent Use: Frequent starting, stopping, reversing can
be possible with the use of control valves
© Provision of Varlable Torque: It can provide variable
torque by use of pressure control valves.
© Explosive Proof: These are generally proot in nature
= Ruggedness: These actuators can be operated in diny,
abrasive wet and mildly corrosive environments.
© Safety: These are safe in operation.
Hydraulic Motor Performance Characteristics
(Theoretical Torque: Due to frictional losses, a hydraulic
motor delivers a torque which is less than its theoretical
value. Theoretical torque is the torque that a frictionless
hydraulic motor would deliver.
1 can be given by following equation =
apa Mee
7” an
Tru = Theoretical torque in (N -m)
Va = Volumetric displacementin
(ren
p = Pressure in (Pa)
Thus, the theoretical torque is proportional to pressure
and volumetric displacement.
(H) Theoretical Power: Theoretical power is the power a
frictionless tycraulic motor would develop, can be
expressed as,
Va p-N
Pe
where, V = Theoretical power in (47)
Va = Volumetric displacement of fuid
in (n*/rev)
p ~ Pressure in (Pa)
N = Speed in (rad/seq)
© scanned with OKEN Scanner RONOTICS (BE MECHANICAL)
(UD Theeretical Flow IRate: The theoretical flow rate is the
flow rate a hydraulic motor would consume if there were
Qn = Va-N
Qha ~ Theoretical flow rate in (m*/s)
Vg - Volumetric displacement in
(oreo)
N = Speed in trevisee)
(lv) Volumetric Efficiency: It is denoted by ny and can be
defined as the ratio of the theoretical flow rate, that motor
should consume to the actual flow rate consumed by
metor,
le ne
ue
where, Qha — is theoretical Now rate (m/s)
Qu ~ is actual flow rate (m/s)
WV) Machanleal Efieleney: It it denoted by nq,, and can be
defined as the ratio of the actual torque delvered by
motor to the torque that motor should theoretically
dllive.
le tet
Th
where, Ta = isthe actual torque (Neen)
Toa ~ ithe theoretical torque (Neen)
[teeter trae
1 * £3)
7
vl) Overalll Efficiency: Overall efficiency can be defined as
the ratio of the actual power delivered by motor to the
actual power delivered to the motor.
itis denoted by ny.
‘Actual power delivered by motor
Te * ‘Actual power delivered to motor
TaN
"a p-Q
where, Ta = Actual torque (N-m)
N = Speed (ps)
p - Pressure in (Pa)
(Q, — Actual flow rate (m'/s)
Here, actual power delivered to 2 motor by the fluid is
called Hydraulic Power and the actual power delivered
to a load by a motor through a rotating shaft is called as
Brake Power.
le
us
DRIVES AND CONTROL SYSTEMS
44.2 Pneumatic Actuators
© Pneumatic devices make use of a fluid medium that is
highly compressible. Fluid is usually air and it is both
readily available and non-flammable,
+ Preumatic actuators ar of fighter construction.
‘These are of nwo types:
Linear actuators - Preumatic cylinders and
(@ Rotary actuators ~ Piston motor or vane mator,
) Unear Actuators
‘These actuators are the means of converting the air pressure
eievered by the air circuit into applied force and straight line
‘motion. Most often used configurations ane a linear single or a
double-acting piston actuator.
Classification:
[Benet om bret] = Banad oe
eae leaner body seeaeaton oe
I T
fesse mang cee ||) ros cptnaer 7
rocemmrnnanie || Reentry | Sm yeaa
icastngactem |ltaonepaensnces || Maton stots
com eee
te Sete ang epinter| |e Meas end pnder
Following Fig. 48 shows constructional details and working of
lunear actuator.
a
Fig. 4. Linear Actuator
* In this case, fluid és compressed air and they typically
operate with pressure at about one hundred pounds per
square inch.
* There & linear motion of gripper attached to piston rod
during retraction stroke. And stroke length is shown in
emended position.
There are two relationships ie the piston velocity of the
‘acuator and the force output of the actuator with respect to
the input power. They are different for extension and
retraction strokes.
© scanned with OKEN Scanner ROBOTICS (BE MECHANICAL) us
For Extension Stroke:
(Forcele = Pressure x Piston area
(velocy)y = Mbubfiow rate
Piston sre
For Retraction Stroke:
(Force), = Pressure x [Piston area - Rod area]
———Input floes rate _
(elec = Ft wee Rod
(0) Rotary Actuator
Usually, air motors are the rotary actuators in pneumatic
system.
‘They convert the power supplied by pressurized air into rowry
motion. Preumatie motors can deliver wiry high rotabonal
‘speed upto 9000 pm or even more.
‘There are two types of rotary actuators:
(Low speed piston motor and
(i) Variable speed vane motor,
1M) Low Speed Pitton Motor: Fig. 49 shows working of
piston motors.
fig. 43: Piston motor
= These motors can be used to provide a smocth source of
Power. They are not susceptible ta overiosd damage.
They can be started and stopped very quickly and with
Pressure regulation and metering of flow can provide
Infinitely wariable torque and speed.
The fire-cylinder piston design prowides even torque at all
speeds due to averiap of the fire power impulses
‘occurring during each revolution of the motor. At beast
{wo pistons are on the power stroke at all times.
= The smooth power flow and accurate balancing make
‘hese motors vibrationvess at af! speeds
©The piston motor has relatively ite exhaust noise, and
this can be further reduced by use of an exhaust ruff.
It is suitable for continuous operation using 100 psi air
pressure and can deliver upto 15 HP.
ll) Variable Speed Vane Motor:
DRIVES AND CONTROL SYSTEMS
© Bconsists of four vanes and it provides rotation in one
Girection only.
©The rotating element is a slotted roter, which is mounted
on a drive shaft. Each slot of the rotor is fitted with a
freely siding rectangular vane. The rotor and vanes
enclosed in the housing, the inner surface of which is
offset from the drive shatt axis.
+ When the rotor rotates, the vanes tend to slide outward
due to centrifugal force. The distance the vane slides, is
limited by the shape of the rotar housing,
‘The motor operates on the ifferential area principle.
When compressed air enters into the inlet port, its
pressure & exerted equally in all directions, Each vane in
tum takes positions and the rotor tums continuously,
© The pressure energy of the compressed air is thus
comwerted into kinetic energy in the form at rotary mation
and force.
© Finally, the air at reduced pressure it exhausted to the
atmosphere
* The shaft of the motor is connected to the unit to be
actuated.
Applications:
* Due to coo! running operations, these motors can be
sed in ambient temperature upto 250°.
+ Ako these are used for mixing equipment, conveyor
rives, food packaging, hoists, tension devices and tum
tables etc.
Advantages and Disadvantages
Advantages:
= Low Power to Welght Ratio: Rotary actuators
employed where lightness and compactness is required
and develop more power per kg weight as compared to
other actuators,
'® Shock and Explosion Proof: Rotary actuators are shock
and explosion proof. As compared to electric actuators
which need very special and costlier construction, these
actuators provide better performance.
© Less Heating: These actuators can be overioaded or
stalled without burning on the other hand the harder the
rotary actuator works, the cooler it nuns. This is due to fact
that air creates a cooling effect while it expands in motor.
© Acceleration and Deceleration Is Fast: Due to low
inertial of rotary actuators, it can accelerate and
decelerate faster. AS compared to the electric actuator,
the rotary actuator does not have the loads on the shaft.
© Glean Operation: As compared to hydraulic actuators,
the leaks from motor can result in damage to material,
which is being processed. It is not there with rotary
© scanned with OKEN Scanner ROOTICS (BE MECHANICAL)
actuators, thus these actuators are found widely
applicable in food processing industry. Also rotary
actuators are easy to maintain due to deanliness.
(© Variable Speed: Using a simple flow control valve, rotary
air motor speed can be waried over a wide range from
4100 rpm to 30,000 rpm.
Disadvantages:
* Nolte In Operation: The air exhaust from the motor
Creates noise, unless itis absorbed by use of muffiers,
‘+ Non-Preciston Metlon: In certain applications, where
high precision motion i needed, rotary actuators are not
Suitable due to compressibility of ai.
Thus, ait actuated systems are highly compliant because
of the compressibility of air
4A} Electrical Actuators
excellent controllability with a minimum of maintenance
required.
‘= There are variety of motors in use for robotic applications,
the most commonly used motors are —
© OC seremotors
(Stepper motors and
Gil) AC servomotors
() Brushless DC motors (BOCM)
(1) DC Servometors
OC servomotors convert electrical energy into mechanical
energy by developing suitable torque on the motor shatt
They have high torque to volume raties. Using dosed
Hoop serva controls, DC motors can be made sumable for
high precision
They provide dean drives as compared to Pycraulic and
Pneumatic actuators.
There are two kinds of motors
© Permanent magnet motors.
Gi) Motors with wound field coils.
Fig. 4.12 shows DC servomotors with wound field coils,
un
Electrical actuators become more popular due to their
DRIVES AND CONTROL SYSTEMS
© The main components of the de servomotor are the rotor
and the stator,
‘©The rotor includes the armature and the commutator
assembly and the stator includes the permanent magnet
and brush assemblies.
+ When current flows through the windings of the
armature, it sets up a magnetic field opposing the field
set up by the magnets. This results a torque on the rotor.
© AS the rotor rotates, the brush and commutator
assemblies switch the current to the armature 50 that the
field remains opposed to the one set up by the magnets.
Thus, the torque produced by the rotor is constant
throughout the rotation,
* As field strength of the rotor is a function of the current
through i,
Then, for a BE servomotor,
Tut = Kyla)
Tu ~ is the torque on the motor
I ~ Is the current flowing through the
armatur
Ku = isthe motor’ torque constant.
where,
Back-EMF:
‘DC motor with armature spinning in the presence of a
magnetic field produces a volage across the armature
terminals, This voltage i proportional to the angular velocity
cof the roter.
te ey (= Kyat
where, ey = isthe backemt
Ky = isthe voltage constant of the motor
@ = isthe anguiar velocity
‘The effect ofthe back-emf is to act as viscous damping for the
motor ie. 2% the velocity increases the damping increases
Propontonately.
Mhertration:
Vin = a voltage supplied across the
motor terminals
Ra = resistance of the armature
2 The current through the armature
Vio
LER
‘This current produces a torque on the rotor and causes the
motor to spin.
Due to spinning of the armature, it generates back-emf equal
to,
eto = Katy
This voltage need to be subtracted from Viq in order to
‘evahuate the armature current.
© scanned with OKEN Scanner ROOTICS (BE MECHANICAL)
2 The actual armature current
ho =
Re
Characteristics:
+ AS the motor speed increases, the back-emt volage
increases, and the current available to the amature
decreases.
‘This decrease in current reduces the torque generated by
the rotor.
* Alsoas the torque decreases, the acceleration of the rotor
decreases.
‘Speed Torque Characterkstics of Shunt Wound Motor:
' sonea
Soeed
Po
—
Fig as
Speed Torque Characteristics of Series Wound Motor:
wl
I
Curent
Soowe
3a
Fig. 18
‘Speed Torque Characteristics of a Compound DC Maton,
te
‘Speed Torque Characteristics of a Permanent Magnet DC
Motor:
Fig. 46
un
DRIVES AND CONTROL SYSTEMS
Undesirable EHtects of Brushes:
They have limited lifetime and require periodic inspection
and replacement.
+ There can be problem of arcing at brushes in an explosive
environment.
* Costof motor with brushes is higher,
+ Brushes cause electrical transients that are a source of
elecvomagnetic interference,
Gi) Stepper Motors
+ These are also called as ‘stepping motors’ It can be
considered as a digital device that provides output in the
form of discrete angular motion increments,
+ tris actuated by a series of discrete electrical pulses. For
every electrical impulse there is a single-step rotation of
the motor shaft
Stepper motors are usually employed for tight duty
‘robotic applications.
+ Ako thete motors are used in open-loop system rather
than the closed oop system,
* A typical application is positioning a work table in two
dimensions for automatic dnilling
+ Fig. 4.17 shows a schematic representation of one type of
stepper motor.
+ Ie consists of a stator and a rotator. The stator is made up
of four electromagnetic poles and the roter is a two-pole
permanent magnet.
1+ W the electromagnetic stator poles are activated in such a
way that pole °C is N (Le. magnetic north} and pole ‘A’ is
S then rotor is aligned as shown,
+ W the stator is excited s0 that pole 'O' is N and pole 'B! is
S, the rotor makes 2 90° turn in the clockwise sense.
Thus, by rapidly switching the current to the stator
electronically it is possible to make the motion of the rotor
en
~=4-f}fE
i
Fig. 4.17: Stepper motor representation
* By determining the number of poles in the stator and
rotor, the resolution of a stepper mator can be fount out.
‘The relation between a stepper motor’s resolution and its
Step angie is given by,
cr
where, R — isthe resolution
and 0 - isthe step angle
© scanned with OKEN Scanner ROBOTICS (BE MECHANICAL)
The relation between a stepper motor’ speed and torque
is not a straight-fine.
The torque is a “function of the angle between the stator
aNd the fotor poles. The torque is maximum when the
oles are aligned, and this value of torque is termed as
the Holding Torque of the motor.
Half Stepping or Micro-Stepping:
his technique to increase the resolution of a stepper motor,
‘thUS reducing the holding torque value
* IN case Of some motors, the speed-torque relation
egrades at certain frequencies of operation and at these
frequencies the operation of the motors must be avoided.
Advantages:
% These motors can be directly compatibie with digital
contrat techniques oF It can be interfaced with digital
controllers
# gives excellent posttioning accuracy.
* Errors associated with these motors are non-cumulathve
ln nature
* AS open-loop control can be used with moter, it is not
Required to use a tachometer and encoder. It results in
considerable cost reduction.
* Motor construction is simple and rugged.
* Motor has a long and maintenance-tree ilfe. Therefore,
tis a cost-ettective actuator.
* The stepper motor can be stalled without causing
damage.
‘Characteristies:
* The motor can be operated In an open-loop way with a
positioning accuracy of # 1 step.
* The motor exhibits high torque at small angular
velocities.
* The motor also exhibits a large holding torque with a
dc excitation. Thus, it acts as a ‘setf-locking’ device, when
the rotor is stationary
* Ade holding characteristic of stepper motor make it an
feremely attractive choice in the actuation of 2 robot
joint.
* Under the supply of full voltage, the motor will move one
step and hold, thus no runaway condition will exist, anct
the stepper is seen to be safer under this type of failure.
Disadvantages:
* Position errors can be caused when one attempts to
move the rotor too rapidly, which i called as
Phenomenon of ‘dropped steps’, and it will couse the
robot arm to reach a final position that is in error.
* The stepwise mation can excite significant manipulator
oscillation. As there is no velocity feedback used, the only
ey
DRIVES AND CONTROL SYSTIMS
way to improve the response is to employ a much more
claborate controtier that is capable of mlerostepplag, It
will raise the cost of the equipment.
iD) AC Servomotors
+ These are cheaper to manufacture,
* They have no brushes and they possess a high power
output
With 2 proper provision of electronics package, the
Performance of AC servomators can be made similar to
that of DC servomotors,
(w) Brushless OC Motors
* Brushless OC motors can be viewed as an “Inslde-Out"
version of standard dc servomator.
Rotor of brushless de motor contains the permanent
‘magnets, whereas the stator consists of the coil segments
aed iron,
* There is no Mechanical Commutation of the coils in
BOCM due to elimination of the Brushes and commutator
ban. A provision of property energizing the stator coil
‘segments must be provided. It is accomplished by placing
lnside the motor itself solid-state devices Le, Hallettect
bipolar sensors that will determine the actual position of
‘the magnets as the rotor turns.
+ A simple electronic logic circuit then processes the
information provided by these sensors, thus, enabling the
‘appropriate stator coil to be excited
# Fig. 418 shows constructional details of brushless. DE
‘ig. 4.1: Brushless DC Motor (8OCM)
Advantages:
4 They require reduced maintenance.
* Brush arcing is eleminated.
Disadvantages:
* They are more expensive
* Rotor inertia is greater than other motors due to the mats
of permanent magnets.
© scanned with OKEN Scanner RONOTICS (BE MECHANICAL) (a1) DRIVES AND CONTROL SYSTEMS
INTRODUCTION TO MANIPULATOR
CONTROL
‘The purpose of robot arm control is to maintain the dynamic
fesponse of the manipulator in accordance with some
prespecified performance criterion. The control requires the
knowledge of the mathematical model and some sort of
inteligence to act on the model
* The mathematical model is obtained from basic physical
laws governing robot dynamics and associated devices.
The intetigence requires sensory capabilities, The
prespecified sk of the robot may be divided in two
lasses, “Contact type task" and “Non-contact type task’,
+ Comuct type task is one in which force or torque
imteraction is involved. For example, a robot arm used for
Spray painting operation, has to grip the spray bottle at
start and then keep it moving along a prespecified
trajectory. Thus, force interaction is not involved in its
operation.
+ Om the other hand, fit is required to scrap the paint trom
2 Glass panel and wipe it clean the end effector, the
scrapping tool in this case, 16 not to be simply moved
along the trajectory, But it must apply a calculated force
against the glass panel
Thus in this case end effector interacts with the
environment. i continuously applies a constant force
against the glass while maintaining its trajectory.
Fig. 4.19 (a) and (b) show both the task listed above.
‘Serny otse
Cane be painted
(a) Non-contact type
_EA ‘abe
(8) Contact type
Fig. 4.19: Types of tats
‘What Is Control?
To control means to regulate, to direct or to command.
Hence, a control system is an arrangement of different
Physical elements connected in such a manner s0 as to
regulate to direct or to command the rabatic system to
Perform the prespecified task.
© scanned with OKEN Scanner RONOTICS (BE MECHANICAL)
Linearity and Non-Linearity of Math Functions and
Controls:
‘There are two types of control systems, namely “Linear control
system” and "Non-linear control system”. Here, iis important
‘To note that Linear control system is not to be confused with
straight line equation oF a linear trajectory. Further every
straight line equation is not mathematically 2 linear function.
‘A mathematical function is ssid to be linear il superimpesition
principle applies to it That is,
it a + by = fa) + fb)
and aay = aKa)
Letus take the function
fy = 3x
Fore = 3
toy = 9
Fork = 5
foy = 15
Now f+ Sp = AG) #24
and also 13) +S) = 24
Tus 1+) = 1G) A)
Also, 12-3) = (6) = 18
and 2-4Q) = 2-938
Thus, 12-3) = 2-10)
Hence, the function f() = 3x i linear.
Now, let us change the function slightly as,
fey = 32-2
Fork = 3
ty = 7
Fore = 5
foy =
Fore = 8
(6) = 22
Bur a) * 5) = 20
Thus, 18 +5) # 1) * HS)
and also (2-3) = 6) =16
2-1G) = 2x7 214
12-3) # 21)
Hence, the function 3) = 3x = 2 is non-finear.
In practice most of the functions are non-linear in. nature
Furthermore, for control systems, if the system under study
‘can be modelled by a linear differential equation, for example
‘a spring-mass system shown in Fig. 420, then itis said to be a
linear control.
and
Fig. 420: Spring mass system
cay
DRIVES AND CONTROL SYSTEMS
mth = 0
We will go into detail of such system later in this chapter.
Fig. 421: Basic control block diagram for robot manipulator
7 CONTROLLER SYSTEMS:
‘The function off controller is to examine the error between set
point and measured value of physical parameter and then
Getermine the action to reduce that error. Thus input to the
‘controller is an error signal. This error between set point value
(expected) and measured valve (present) is detected by the
‘error detector, The output of eor detector is an error signal
which is then applied to the controller, Note that error
derecior finds error only when set point value {i5) and
measured value (m) are in same form, Let us see some
imporiant terminologies that are required in further
discussion.
ATA Error
The error or deviation of the controlled varlabla from set point
is gen by,
erred (4d
where,
= Error in absolute value
b= Measured value of variable
1 = Setpoint of variable
(reference)
Generally, error is expressed in percentage of span of
measured value.
«100 (42)
b
OF na Bond
4.7.2 Control Parameter Rang:
This range is assodated with the controller output. The
controfer output as a percent of fullsale is given as:
Uae
nn Une
where, P = Controller output as percent of
ful scale
Value of output
ee = Maximum value of controling
parameter
Ue = Minimum value of controlling
parameter.
Pe x 100 (43)
© scanned with OKEN Scanner RONOTICS (BEMECHANICAL)
4.1.3 Controller Modes
‘The controler produces a output (1e. control signal) to final
control element based on input error signal (Le. set point —
‘measured variable). The objectives of controller are,
(2), Minimum deviation/error and
(0) Minimum durationtime.
‘©The measured value should reach to the set point value
within a small time. How it is achieved by the controller ?
‘The answer is : Controller solves some mathematical
equations to calculate its output (ie. control signal). These
‘mathematical equations are called “controller modes” or
“controller actions".
© There are many mathematical equations available 10
calculate controller output. All these equations are not
required every time. Which equation is suitable for our
application? Selection of mathematical equation (ce.
Controller made) depends upon process characteristics.
‘The process characteristics are well explained in chapter 4
These are:
(2) Process equation
(0) Process load
(© Process lag
(@) Self regulation
(©) Capacity of the process
CControlier modes show how controller responds to the error
‘Signal. These equations are also called ‘control strategies’. To
‘tolve these mathematical equations, we need some hardware,
‘According to the hardware used, the controllers are classified
4a follows,
4.74 Types of Controllers
‘The controllers are broadly classified according to system
used as:
2. Preumatic controllers
2. Hydraulic controllers
3. Electronic controllers
‘The flapper-nozle system is building Block in pneumatic
controller to solve some mathematical equations. Hydraulic
controller uses jet valves and pistion-cyfinder arrangement for
‘the same.
Electronic controllers are classified into =
(0) Analog Controllers : Operational amplifier (op-amps) is
the main component.
{b) Digital Controtters Above equations are solved using
sofware in computers, programmable logic controllers,
microprocessors, etc
(ann
DRIVES AND CONTROL SYSTEMS
‘The purpose of above different types of controller is to solve
‘only mathematical equations (control actions). Type of
‘controller is important specification of a controller. But which
‘equation is suitable for application does not depend upon :
(2) Type of eontrolters,
(©) Type of processes.
Selection of suitable control action depends only upon
process characteristics
‘Types of Controller Modes
Fig. 4.22 shows oitlerent controller modes,
In discontinucus control mode, the controller output changes
in discontinuous manner with the «error input, There Is. a
‘certain controlier output for some band of error. If an error
‘occur is more than this band, then onty controller output is
changed stepwise. These controllers are simplest, cheapest
and form the basis of continuous control mode.
There are three types of discontinuous control mode name!
Two position (ON-OFF), Three position (multiposition) and
time proportionating (Floating) control mode.
‘Conte moses
——,
Fig. 4.22 Types of Controller modes
* In continuous control made, the controller autput
changes in smooth manner (continuously) with the error
input They are more accurate, reliable, effective on
isturbances and suitable for complex process. For every
‘value of error, there is certain controller output. There is
One-to-one relationship between error and controller
ouput
* According to relationship between emor and controller
output the different types of continuous control modes
are : Proportional (P}, Integral (I) or Reset, Derivative or
Rate (D). Each mode having some advantages and
drawbacks.
+ To collect the advantages of individual mode and to
eliminate its drawbacks, these modes are combined
© scanned with OKEN Scanner RONOTICS (BE MECHANICAL)
(ann
together which is called “combinational or composite
Control mode". It includes Proportional Integral (P + 0,
Proportional = Derivative (P + 0) and Proportional
Integeal Derivative (P + 1 + 0) controller mode
‘Continuous Controller Mode
In these modes, the controller output changes continuously
with the error input They are more accurate, reliable and
«effective on disturbances and suitable for complex processes
‘The types of continuous comrol modes are according to
Felationship between controller cutput and error input. They
are:
12) Proportional (P) control
(O) Integral or Reset control
(0) Derivative (0) or Rate or Anticipatory control
(@) Combinational or Composite control made
(@ Proportional + integral (P+
Gi) Proportional + Derivative (P+D)
Gi) Proportional + tntegral + Derivative (P=1eD)
Let us see every mode in detail
4.7.5 Proportional Control (P)
In proportional contrel mode, the controller output (?)
‘changes linearly with the error input (¢,) Linear relationship és
given by a straight line equation y = mx * ¢ for x input and y
‘output. Similarly, matnematcally, proportional controller is
represented 25:
Pm Kee Pe
P = controlter output — (%)
t= erroringt— (%)
Ky * proportional gain — (/%)
P. = controller output when error = O__ (%)
‘Above equation (4.4) represents reverse action of contrater, in
\which when error is negative, controller output is decreasing
For direct action,
—84)
where,
Pn = Kee Py -45)
435 Integral Control Mode)
Tis also called “Reset controller because it resets the offset
{error}, which is produced in proportional contro! mode when
2 disturbance occurs. Integral mode produced zero-eror
‘eutput eventhough a disturbance occurs. In this mode,
measured value approaches to setpoint value with no error.
In floating control mode, the controller output changes with
‘some fixed or multiple speed depends upon err zero. But in
Integral control mode, the controller output changes
continuously with the error. It is defined as, “Rate of change of
controller output is proportional to the error. Mathematically,
itis given by equation (46),
es
a
45)
DRIVES AND CONTROL SYSTEMS
where, 2 = ate of change of contrller
output... 06/58)
= Exot %
‘The equation (46) shows that there is continuously change in
rate of output until error becomes zero. Thus it resets the
‘error, I the error is large controller output changes with large
fate and when error is less, the output changes with small
rate.
'* Condusion of P, 1 and D actions : Table 42 shows
response of P, and D controller for standard input signal,
' JY
t aL
4.7.7 Composite Control Modes
The control modes discussed up til now are not generally
Used alone because they have certain limitations as shawn in
Table 41. So to collect advantages of each mode and
eliminate its drawbacks, these modes are combined together
which are called ‘composite control modes’. They are :
{a) Proportional - Integrat Control Mode (P-N).
(©) Proportional = Dernative Control Mode (PD).
(€) Proportional = Integral ~ Derivative Control Mode (P-I-D).
Let us see each mode in deta
Proportional - Integral Control Mode (P-1)
‘TWis mode is formed by the combination of proportional and
integral control modes. Mathematically It is written as :
t
Pe meeeiem Seater, 47
°
where, K, = Proportional gain
K = Integral gain
P= Integral term value att = 0
This mode collects advantages 2s :
(2) One-to-one correspondence between Pand ¢,
(©) Eliminates offset produced by P-mode
I may have drawback of sluggish response. This is very
Popular control in industry. More than 95% applications
required Pl controller. In proportional mode, offset is
produced when a load change (disturbance) occurs, offset
is fed error state which is eliminated by integral term.
© scanned with OKEN Scanner RONOTICS (BE MECHANICAL)
‘© InP controller, when error is zero, the controller output is
fixed at the value that the integral term had, when error
went to-zer0. Then P= P,
© IW the errar is not zer0, the proportional term contributes
2 correction and integral term begins to increase or
Gecrease the accumulated valve.
‘© The integral term cannot be negative. Thus, it will saturate
to zer0, This controller has sertings of Ky hand P, For
better control, Ky should be large (smal Py. Ky should be
lange (small T) and P= SO%
Repeats Per Minut
This terminology is related to integral term of PL controler,
‘The gain K, has the effect of causing the controller output to
change every unit time by the proportional mode amount
Taking derivative of equation (47,
OP = Kika oat 48)
Here, K€ = Proportional contribution
St = Unit time interval
and X= Repeats the proportional gain per
minute
" = 05%
and 10%
then ss
" 10% f Semin then for every
minute the output is increased by
S% times LOK/A-min oF SOX oF
10 repeats per minute,
Ik repeats the proportional amount 10 times per minute. The
‘umber of minutes I term take to repeat action of peter.
When to Select P-I Controtier:
Pl control mode is selected for the processes huving
following characteristics:
(a) Frequent or large load changes (disturbances)
fb) Moderate process lags
(0) Fast reaction rate
(0) Any capacty process
Proportional-Dertvative Control Mode (P-D)
‘This control mode is result of cambination of proportional
«and derivative control modes. Mathematically, its written 2s:
P= Kee kK at P -43)
where,
aan
DRIVES AND CONTROL SYSTEMS
Advantages of these Nodes are:
* One-to-one correspondence between Pand ey
+ Rapid inital response, which handles fast load changes.
But as there i no integral action, the drawback ist cannot
‘eliminate the offset of proportional action, This controller
used where offset error is acceptable. Because of derivative
ction, overshoots and undershoots are reduced and it takes
less time to get steady-state response. Let us see some
Examples
Example 4.1: Drow the controller response for given error
(graph os shown in Fig. 5.23 K, = 5, Ky = 0S 0nd P, = 20%,
Fig. 423: Error input to controller for Example 4.
Solution: The proporional-derivative controller output is
given by
P os
Bt Kot Pe
(Over three ranges of error :0 to 1 sec, 1 to 3 sec. and 3t0 5
see.
(a) Forerror range 0 to 1 second,
nat
Pye Kyeate KyKorat Py
Here a 2 1%
* Pls Ste25+20
ate, 225%
ated 25%
@) For eror range 1t0 3 seconds ey = 1, a.
Pre 595 (05)x0420
P, = 25%
From t = 1 tot = 3, controller output = 25%,
(©) For error range 3 to 5 seconds,
e, = 051425
S(-OSt+ 25) + 5(05)(-05)+ 20
= 250+ 125-125 +20
= -25e+ 31.25
‘These outputs wrt time plotted are as shawn in Fig. 4.24,
Then
© scanned with OKEN Scanner Fig. 424: Controller response for Example 4.1
When to select P-D controlier :
P-D controller is suitable for the processes having following
process characteristics:
1G) For fast load changes.
(0) Forlarge process lags.
(0) For small or urge capacity process.
(0) For the process where tse generated by the
proportional control mode is tolerable.
(e). For process having large process equation.
Proportional-Integral = Beehvathve (PID) Control Mode
This is one of the pawerul but complex controler mode
combining proportional, integral and derivative control
modes. It has following advantages :
a). One-to-one comespondence between P and ¢,
(0) Eliminates offset of P-mode.
(0) Provides fast response.
Mathematically, PID controle is written as:
Pa ee KK,
+ ye Ee DD
Kp = proportional gain 5%
Ky = integral gain —_%/secr%
Ko = derivative gain sec
K, is expressed in terms of proportional Band Py
jis expressed in terms of integral time T, and
Ky is expressed in terms of derivative time To.
For better control Py should be less, T; should be less and T3
should be high.
PID controller can be used as P controller by removing | and D
actions. Integral action can be removed by using
7, = maximum. Derivative action can be removed by using
To = O. Similarly PID controller can be used as Pl and PD
controller, by removing D-action (Tz = @) and Lacon
{= maximum) respectively.
where,
(aa)
DRIVES AND CONTROL SYSTEMS
‘When to Select PID Controller:
PID controller is suitable for the processes having following
chancteristics:
(2) Large process load changes,
(©) Large process equation.
(c)_ Large process lags eg. Temperature process.
(@) Any capacity process.
Let us see the changes in transient responses by adding I and
Daction in P-controller.
P40 ncton
@
Fig. 42%: Contratier sponte (a) p-action, (b) P-f action,
Ke) PED action,
SECOND ORDER CONTROL SYSTEM
For a second order control system, relationship between input
Parameter x and output parameter y is described by a
Gifferential equation of the form
coendeen tiene
—
Fig. 26: Spring mass system with surtace friction
Let us tate an example of simple mechanical system, spring
mass system shown in Fig. 426.
# AbBlock of mass m' is attached to a rigid support through
a spring having stiffness "k. Frictional resistance to the
movernent of the block is assumed to be proportional to
the velocity of block.
+ There are in all three forces acting on the block one is due
10 acceleration, other due to spring stiffness and the last
due to friction.
© scanned with OKEN Scanner ROROTICS (RE MECHANICAL)
* Ablock of mass'm' is attached to a rigid support through
Spring having stiffiness Frictional resistance 10 the
‘movernent of the block is assumed to be proportional to
the velocity of block.
* “There are in al three forces acting on the block one is due
to acceleration, other due to spring stiffness and the last
due to friction.
Hence, the equation of motion becomes,
me sbt six = 0
31)
+ As the coefficients are constant and degree of the
equation is 1 the differential equation is linear. any one
(OF more of the three characteristes are non-linear the
differential equation would become non-linear. For
example, if spring has non-linear stiffness characteristic as
Force = qx? instead of Force = lex (where, @ and k are
Constants Gf proportionality) the equation’s degree will
change and it will become 2 noo-tinear clifferential
equation. Here, we will restrict our scope to discuss
Solution of linear differential equation only. The selution
Of above equation is a time function a(t) which describes
mation of the block
© One must remember that (0) and i{0) both cannot be
‘equal to.20r0, One of the two parameters of both must be
on-2ara, to disturb the system and set it into motion.
‘Thus, the system is set in motion because of initial
isplacement and/or initial velocity. Ifthe spring has very
less stiffness but surlace has greater friction the block
‘would return to its resting position in a very tlow,
sluggish manner.
* But if the spring is very sti and the surface is very
smooth, the block would oscillate several tmes before
ataining the state of rest. Ths, diferent possibilities arise
because of the characteristic values of m, band & Hence,
the equation cf motion may be reveitien as characteristic
equation
me ebsek= 0 =a
+ So that roats of this quadratic equation will help us in
getting solution to the differential equation.
‘The two roots of the quadratic equation willbe,
cbs Yo!—amk
as 7
fa? — ark
and 2m
2
aay
DRIVES AND CONTROL SYSTEMS
(Due to discriminant b? — 4mk, there are three possibilities as
follows:
1 The Real and Unequal Roots: fb? > amk, s, and 5; will
be real but not equal to each other, Physical significance
of this case is that the friction is dominating characteristic,
hhence the motion is slow and sluggish and system is said
to be Overdamped. The time function for the motion of
block then becomes,
wo + Gell cel a4)
2 Complex Roots: Ib? < Amik, roots 5, and ¢; will have real
and imaginary pans and they will be complex conjugates.
‘The solution of differential equation in this case although
Same as in previous case, is dificuR 10 use directly since it
involves imaginary numbers explicitly, Hence, the solution
{5 manipulated using Euler's formula given below.
eins tit
= (0058 + isin)
This manipulation of the solution will be more clear after
studying a solved example given at end ot this section.
Physically, the block oscillates before it reaches a steady,
Fest position, due to dominating stitiness characteristic,
and the system ig said t0 be Underdamped.
3. Real and Equal Roots: This ta special case when
bP = demk, that is friction and stiffness are “balanced”
Welding the fastest possible non-oscillatory motion of the
block. The system is said to be Critically Damped.
ln this case form of the time function for motion of the
block ix,
x9 = Gee ce’
AS esas
MO = C+ cnet (4.15)
48.1 Control Law the Second-Order System
1 the natural response of the system shown in Fig. 4.26 not
25 per our requirement, then an external force may be applied
Con the bloc 40 that the system will respond as we wish, That
is when the system has certain values of parameters mb and
kas seen in previous three solved problems the system will
have cenain faued natures of response namely overdamped,
underdamped and critically damped. If, for given values of m,
‘DB and k system has underdamped response but we wish to.
bring it to the rest as fast as possible by critical damping, then
without actually changing the spring and without having
control over the actual frictional force offered by the surface,
wwe can apply a calculated extemal force on the block, so that
‘wal respond lke a crteally damped system.
© scanned with OKEN Scanner ROBOTICS (BE MECHANICAL)
© To understand this process let us take 2 general example
which is familiar to us ang will make the concept clear
Imagine a bicycle moving on a horizantal road with initial
velocity Vit we do not pedal the wheels, the bicycie wall
come to a halt, automatically, after certain time Y. Lets
say that this is natural response of the system. Now, if we
wish to bring the bicycle to a hatt in 1/2" time period, then
physically we cannot change the frictional resistance
offered by the road, but we can provide equivalent
frictional force extemally by applying brakes In other
words, we are gaining @ frictional force separately, which
adds to the frictional resistance offered by the road.
Determining this “Gain” in frictional resistance is known as
establishing a “control law”
Now, coming back to the spring mass system. if we wish
the system to respond in cenain manner then
determining the modified values of parameters b and/or k
is known as establishing the “control law’. In other words,
we will calculate “Gain* in b and ik, which ave
characteristics of velocity control and position contro!
respectively
The spring mass system may be provided with an actuator
to control its behaviour by applying predetermined force
fas shown in Fig. 427. Now, we will discuss the
procedure for getting mathematical expression of
pte
Fig. 4.27: Spring mass system with actuator
‘The ditferential equation for motion will Be,
mettre et (436)
We wish to find modification in b and k such that the equation
‘will become,
mi +b + kx = 0 ~417)
‘and b’ and k will have values such that systems response is as
er our wish, say critically damped For that purpose force
should be,
Fe tox een ap
where, kip is gain in position control oF in other words
additional stfiness characteristics and ky is gain in velocity
Control oF in other words additional friction characteristics
“The differential equation becomes,
me Fekdt orbs = 0
(aan
DRIVES AND CONTROL SYSTEMS
‘The equation (4.18) is the “control law" for achieving certain
behaviour of the system. Fig. 427 is a block diagram to
‘explain now feedback is taken from system and how modified
inputs are fed to the system. (Fig. 428)
f= te - 7x
Fig. 4.28: Closed loop ryster with control law
given by equation (4.18)
‘The differential equation of the system without control law is,
remem eo
Hence, characteristic equation is,
Pemereo
Ater applying control law, the differential equation will
become,
ade tee teat
Thus 1+ Gx +r = 0
Hence, characteristic equation becomes ¢
the system wil become critically damped.
The modified values of frition characteristic (b) and stiffness
characteristic (() are also called as closed loop friction
characteristic and closed loop stiffness because they are
determined after getting feedback of originat behaviour of the
sytem. Thus, the original “open loop” system becomes a
“closed loop” system due to the feedback control
4.8.2 Control Law Partitioning
Wile designing control laws for mare complicated systems
(including non-linear systems} it becomes difficult to have
total control on the system by simply additional gains in
position control and velocity control because these gains are
individually dependent on system parameters.
Hence, the controller, in turn the contral law is partitioned in
‘wo parts as modielstased portion and servo-portion.
Pe 6s+9 = 0 and
@ scanned with OKEN Scanner ROUOTICS (BE MECHANICAL)
‘As discussed earlier the open loop equation of motion for the
‘system shown in Fig. 4.27, is given by equation (4.20) that is,
me tbe she =f — (429)
We call it “open loop” because still feedback was not
introduced in the system
Now one part of the control law, that i, model-based portion
will make use of parameters m, bk and will reduce the system
‘So that it appears to be a unit mass system as follows.
Let feof +s 42)
where, @ and § are either functions or constants chasen in
such a way that if f is taken as the mew input to the system,
‘the system appears to be a unit mass.
By equating equations (420) ane (421)
ma ee okr = Of #8
Riscleartna, = m
and B= broke
so that system appears to be a unit mass, and the system
‘equation becomes,
aee (422)
Now, as the other part of control law, that is servo position
we may have
Fn het ht 2
as we did earlier in equation (422)
Now, Combining equations (4 22) and (423), we get,
Pehle =0
Equation (424) leads to the characteristic equation as,
Petsety =O
which haste two 08s,
— (424)
(425)
4
For achieving behaviour of the system to be critically damped
the discriinar = mast be zero. Tat means,
ky = 2k — (426)
Equation (4.26) makes the gains independent of the system
parameters. Thus, the control law is partitioned to modify the
system parameters @ B and the gains kp. i, separately
Fig. 429 explains the mathod af control law particning.
cap
=] El
Fig. 425: Control law partitioning
48.3 Additional Observations on Partitioning of
Control Law
HM we do not use panitioning methodology, we get differential
‘equation of the system as,
eke BO
which reduces to
Tem oae eo
land the characteristic equation becomes,
Ceieeed
The equal roots are,
eg
Hence, ay = (Ge ne
and. MO = 2G 0 Ce -2G 10"
Using initial conditions as x) = 2, say and x(0) = 0
Gye 2andCze4
Thus, x = Qe ane
Now with the method of parttioning the control taw, the
Gitferential equation becomes,
1+a¥20e8= 0
Hence, characteristic equation is,
FraVis-8 20
The equal roots will be
sen
2 ao =
aed. Moe
= (4.27)
Using the came initial conditions, as x{0) = 2, x(0) = 0
Coe 2and Gea?
Thus, 0 = Qe ayine™ (4.28)
Let us now compare equations (428) and (429) as below,
@ scanned with OKEN Scanner ROUOTICS (BE MECHANICAL) aan DRIVES AND CONTROL SYSTEMS
Table 4.3 where, @ = Emor or disturbance in the desired
Timat | 10 witout parttioning | af wth partticning position
° om a = Desired position ~ Actual position
02 1.8769 Tes "-*
on ters ise xis the time function of end effector’ actual position and xj
os 153 sess is the te function of end effector’ desired position.
on 1069 osm 1 the parameters mb, k are perfectly known and if there is no
7 Bei Gee initial error, the enc effector wil follow the desived trajectory
7 es a ‘exactly. And if there is initial error, then it will be suppressed
bi came Xccording to equation (429) and thereafter the end effector
ma will follow the trajectory exactly, The equation (4.29) must be
3 ihe. eau modified only it there is any external disturbance.
= ir on The disturbance generated outside the system and acting as
“ ame _— ar ees put Been addon oral sk
but affecting the output adversely is called as an external
disturbance Hence, one purpote of the control system is to
come Prove disturbance rejection. f the external disturbance i
‘equivalent to a force fain equation (429) will be rewriten as,
; wn
ats
7
ig. 4.30: Response of the nystem without and with
control law partitioning
‘The sole objective of controlling the system i to bring it to
the state of rest as fast as possible. From the above table and
Fig, 430, it ts clear that the methodology of control law
is beneficial in this sense too
9 TRAJECTORY FOLLOWING CONTROL
Til now, we have discussed how the system can be brought
back to the state of rest as its original position, f disturbed.
Now lets understand what is disturbance with reference to the
lend elfector movement. The end effector is supposed to
follow the trajectory planned by the planner, Thus, the end
‘effector is supposed to be positioned at particular
Predetermined location at given instant of time At that
instant, if the end eHfector is not located at the desired
Position, the difference between the two locations is known as
‘error or the disturbance. The sole objective of the control
system is to minimize this error (or reduce it to zero) as
‘quickly a5 possible Thus, the function x(t) discussed earier
may now be treated as equation written in “error space” oF
‘equation of motion to minimize error, Therefore, equation
(6.54) may be rewritten 2s,
er kerke = 0 429
yerkoe © fast (430)
The disturbance may be of any nature like varying with
respect to time which may be modelled with certain time
function, or it may be constant
To understand the methodology of disturbance rejection, let
us consider a simplest kind of disturbance, obviously, a
constant disturbance, that is,
feist = constant
\which means error in locating end effector at desired position
@ = constant
Naturally, frst and second derivatives of ‘e* with respect to
time Y e and ¢ are equal to zero.
Hence, from the equation (4 30),
Ripe = feist
and e
‘That implies, higher the position control gain (kp), smaller the
‘error AS errors’ is not zero, the disturbance is nat completely
rejected,
Hence, for disturbance rejection, the control law equation
(430) may be modified by an additional integral term as
follows.
rivertpe sti S edt = lane ay)
‘The integral term is added so that error is not steady although
fai 'S Constant
@ scanned with OKEN Scanner ROBOTICS (BE MECHARICAL) (420 DRIVES AND CONTROL SYSTEMS
Dir iches emierettation: os sepa Se Om = Re speed of motor
erkyeriperke=d — (432) | The armature cireuit of OC torque motor is as shown in
‘With this control law the system becomes third order system
To reject the disturbance completely, solution of differential
‘equation given by equation (437) may be determined. It is
‘obvious to keep integral control gain (k) as small as possible
{to bring the third order system almost dose to second order
system. The control law given by equation (431) is
Proportional-integral-Derivative (PID) control as i includes all
the above mathematical terms.
11 MODELING AND CONTROL OF A SINGLE
JOINT.
aut of a Single Joint
Usually, OC motor is used as an actuator in many industrial
robots. The motor works on the principle of
flectromagnetism. It has non-rotating part that is stator
consisting housing, bearings and permanant or
lectromagnets.
The rotor consists of shat and windings through which
‘current flows, © that tangential force is generated with the
robot which in tun generates torque. The tangential force is
expressed as,
Fe qve
where
a = Charge
¥ = Speed with the charge moves
‘trough the windings
8 = Magnetic field set up by the
magnets in stator
But, this equation won't lead us to model the system by a
ifferential equation
So, considering torque capacity of motor which is
proportional to current flowing through the armature, it may
be stated that,
Tm = kmia
Tm = Torque capacity of motor
lim = Motor torque constant
iy = Armature current
Funher, when the rotor rotates, it generates a small amount of
electromotive force across the armature, as it rotates in the
magnetic fied. Ths electromotive force (emf) or the voltage is
Proportional to motor speed. Hence,
V = kab
— (433)
where,
(434)
where,
V = Voltage generated
ke = Back emf constant
Fig. 432.
Fig. 431: Armature circuit of OC torque motor
where,
V, = Armature souree voltage
iy = Armature current
V = Voltage generated due to rotation
Ry = Armature resistance
La = Armature inductance
Practically, it is observed that inductance of motor is
negligible, hence the circuit may be represented by equatian
Va-V = laa
Tn ‘
Vas ED Rat hem 435)
This i fest order
