Sunteți pe pagina 1din 6

6.

003: Signals and Systems Lecture 12 March 16, 2010

6.003: Signals and Systems Feedback and Control

Feedback: simple, elegant, and robust framework for control.


CT Feedback and Control

E C
X + Y

controller plant

sensor

Last time: robotic driving.

di = desiredFront
do = distanceFront
March 16, 2010

Feedback and Control Op-amps

This week: using feedback to enhance performance. An “ideal” op-amp has many desireable characteristics.

Examples:
• increasing speed and bandwidth V+
• controlling position instead of speed K Vo = K (V+ − V− )
V−
• reducing sensitivity to parameter variation
• reducing distortion
• stabilizing unstable systems • high speed
− magnetic levitation • large bandwidth
− inverted pendulum • high input impedance
• low output impedance
• ...

It is difficult to build a circuit with all of these features.

Op-Amp Op-Amp

The gain of an op-amp depends on frequency. Low-gain at high frequencies limits applications.

audio
105 105
frequencies
|K(jω)| [log scale]

|K(jω)| [log scale]

104 104

103 103

102 ω [log scale] 102 ω [log scale]


1 10 102 103 104 105 106 1 10 102 103 104 105 106
0 0
∠K(jω)|

∠K(jω)|

− π2 ω [log scale] − π2 ω [log scale]


1 10 102 103 104 105 106 1 10 102 103 104 105 106

Frequency dependence of LM741 op-amp. Unacceptable frequency response for an audio amplifier.

6.003: Signals and Systems Lecture 12 March 16, 2010

Op-Amp Check Yourself

An ideal op-amp has fast time response.

Vi Determine τ for the unit-step response s(t) of an LM741.


K Vo
105

∠K(jω)| |K(jω)| [log scale]


s(t) 104

Step response: A 103

Vi (t) = u(t) Vo (t) = s(t) t 102


τ [seconds] 0
1 A
− π2
t t 1 10 102 103 104 105
ω [log scale]
1. 40 s 40 s
2. 2π 1 s
3. 40 4. 2π 1
40 s 5. 2π×40 s
0. none of the above

Op-Amp Dominant Pole

We can use feedback to improve performance of op-amps. Op-amps are designed to have a dominant pole at low frequencies:
→ simplifies the application of feedback.
circuit 6.003 model

Vi s-plane
K(s) Vo Vi + K(s) Vo

R1
βV0 β −40
R2

Vo K(s)
=
Vi 1 + βK(s)
� �
R2 40 rad/s
V− = βVo = Vo α = 40 rad/s = ≈ 6.4 Hz
R1 + R2 2π rad/cycle

Improving Performance Check Yourself

Using feedback to improve performance parameters.


circuit 6.003 model What is the most negative value of the closed-loop pole
that can be achieved with feedback?
Vi
K(s) Vo Vi + K(s) Vo
− s-plane
R1
βV0 β
R2
? −α
Vo K(s)
=
� � Vi 1 + βK(s)
R2
V− = βVo = Vo αK0
R1 + R2 = s+α
1. −α(1 + β) 2. −α(1 + βK0 )
1 + β αK0
s+α
3. −α(1 + K0 ) 4. −∞
αK0
= 5. none of the above
s + α + αβK0

6.003: Signals and Systems Lecture 12 March 16, 2010

Improving Performance Improving Performance

Feedback extends frequency response by a factor of 1 + βK0 Feedback produces higher bandwidths by reducing the gain at low
(K0 = 2 × 105 ). frequencies. It trades gain for bandwidth.

1 β = 0 (open loop)
[log scale]

105
0.1 β = 10−4
1 + βK0 104

|H(jω)| [log scale]


|H(jω)|
|H(j0)|

0.01 103
β = 10−2
0.001 ω [log scale] 102
1 10 102 103 104 105 106 10
0
∠H(jω)|

1 + βK0 β=1
1
− π2 ω [log scale] 0.1
1 10 102 103 104 105 106
ω [log scale]
1 102 104 106 108

Improving Performance Improving Performance

Feedback makes the time response faster by a factor of 1 + βK0 Feedback produces faster responses by reducing the final value of
(K0 = 2 × 105 ). the step response. It trades gain for speed.

Step response Step response

K0 K0
s(t) = (1 − e−α(1+βK0 )t )u(t) s(t) = (1 − e−α(1+βK0 )t )u(t)
1 + βK0 1 + βK0

s(t) s(t) β
K0
2 × 105 0
1 + βK0 β=1
β=0 0.5 × 10−5
1.5 × 10−5
t [seconds] t [seconds]
1/40 1/40

ds(t) ��
The maximum rate of voltage change is not increased.
dt �t=0+

Improving Performance Motor Controller

Feedback improves performance parameters of op-amp circuits. We wish to build a robot arm (actually its elbow). The input should
• can extend frequency response be voltage v(t), and the output should be the elbow angle θ(t).
• can increase speed

Performance enhancements are achieved through a reduction of gain.


robotic
v(t) θ(t) ∝ v(t)
arm

We wish to build the robot arm with a DC motor.

v(t) DC motor θ(t)

This problem is similar to the head-turning servo in 6.01 !

6.003: Signals and Systems Lecture 12 March 16, 2010

Check Yourself Motor Controller

Use proportional feedback to control the angle of the motor’s shaft.

What is the relation between v(t) and θ(t) for a DC motor? amplifier DC motor
v(t) + α γA θ(t)

v(t) DC motor θ(t) β

feedback
1. θ(t) ∝ v(t) (potentiometer)
2. cos θ(t) ∝ v(t)
3. θ(t) ∝ v̇(t) Θ αγA αγ 1s αγ
4. cos θ(t) ∝ v̇(t) = = =
V 1 + αβγA 1 + αβγ 1s s + αβγ
5. none of the above

Motor Controller Motor Controller

The closed loop system has a single pole at s = −αβγ. Find the impulse and step response.

Θ αγ The system function is


=
V s + αβγ Θ αγ
= .
V s + αβγ
ω
s-plane The impulse response is
h(t) = αγe−αβγt u(t)
and the step response is therefore
σ 1� �
s(t) = 1 − e−αβγt u(t) .
β
θ(t)
1
β
α↑
As α increases, the closed-loop pole becomes increasingly negative. t
The response is faster for larger values of α.
Try it: Demo.

Motor Controller Motor Controller

The speed of a DC motor does not change instantly if the voltage Analyze second-order model.

is stepped. There is lag due to rotational inertia.


amplifier DC motor

First-order model Second-order model γpA2


v(t) + α θ(t)
integrator integrator with lag − 1 + pA
� �
pA
V γA Θ1 V γA Θ2 β
1 + pA

feedback
Step response of the models: (potentiometer)

v(t) θ(t) αγpA2


θ1 (t) = γtu(t) Θ 1+pA αγpA2 αγp
1 � � = = = 2
θ2 (t) = γt − γp (1 − e−pt ) u(t) V αβγpA2
1 + 1+pA 1 + pA + αβγpA2 s + ps + αβγp
�� �
t t p p 2
s=− ± − αβγp
2 2

6.003: Signals and Systems Lecture 12 March 16, 2010

Motor Controller Motor Controller

For second-order model, increasing α causes the poles at 0 and −p Step response.
to approach each other, collide at s = −p/2, then split into two poles
s(t)
with imaginary parts.

ω 1
s-plane β

σ
−p

Increasing the gain α does not increase speed of convergence.

Motor Controller Motor Controller

Step response. Step response.

s(t) s(t)

1 1
β β
e−pt/2 cos(ωd t + φ)
t t

Feedback and Control: Summary

CT feedback is useful for many reasons. Today we saw two:


• increasing speed and bandwidth
• controlling position instead of speed

Next time we will look at several others:


• reduce sensitivity to parameter variation
• reduce distortion
• stabilize unstable systems
− magnetic levitation
− inverted pendulum

MIT OpenCourseWare
http://ocw.mit.edu

6.003 Signals and Systems


Spring 2010

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

S-ar putea să vă placă și