國立交通大學電機與控制工程學系 90 學年度第二學期

課程名稱：感測與介面（課號：DCN3701/1123）
作業#2 (Due Wednesday 3/27 10:00AM)
This homework covers the contents of Dynamic Characteristics of sensor devices.

1. Give an example to show that an high gain negative feedback can be applied to a
sensing element for improving the its dynamic characteristics. 15%
Ans：
Unbalanced
force
1 Amplifier
Interia
a m force + k x E i
1 2 2ξ KD KA R Vout
- s + s +1
seismic Electro- ωn
2
ωn Potentiometer
Standard
mass magnetic displacement
Elastic force resistor
force sensor
sensor

Kf
i
Coil and
magnet

One example is the closed-loop accelerometer, and its block diagram is shown above.
The overall system transfer function is:

 
~  
∆ V ( s ) mR  1 
=  
~
∆ a( s) K F  k 1 2 2ξ ks  k 
s + + 1 + 
 K A K D K F ω n2 ωn K AKD KF  K AKD KF  

If K A → ∞ ，so that K A K D K F k >> 1，the system transfer function can be reduced as
~
∆ V (s) Ks
=
~
1 2 2ξ s
∆ a( s) s + s +1
ω ns 2
ω ns

where
mR
system steady-state sensitivity Ks =
KF

K K K 
system nature frequency ω ns = ω n  A D F 
 k 

 k 
system damping ratio ξ s = ξ  
 K AKDKF 
⇒ We see that the system natural frequency ω ns is now much greater than that of

the elastic force element itself. The system damping ratio ξ s is much less than ξ ,
 but by making ξ large a value of ξ s ≅ 0.7 can be obtained. Furthermore the system

steady-state sensitivity depends only on m, K F and R which can be made constant to a

high degree.

2.
A. Explain what is the meaning of bandwidth of a sensor. 10%
B. How to determine (or sometimes estimate) the bandwidth of a first order
element 5%
C. How to determine the bandwidth of a second order element with damping ratio
< 0.7? 5%
Ans：
a)

數學意義：頻寬(Bandwidth)意指一系統或一元件之 magnitude ratio 大於 1 範圍

2
內所包含的頻率。
物理意義：頻寬(Bandwidth)意指一系統之響應大小衰減至-3dB 之範圍，一般來
說，頻寬越大，元件操作範圍越廣，反應越快，性能也越好。因此對
於一般的量測系統來說，頻寬(Bandwidth)為一不可忽略之因素。

b)
1
1o 一階系統之轉移函數為 G ( s) = ，其中τ為時間常數(time constant)。
1 + τs

2o 求初期響應衰減至 1 時的頻率，我們將 s=jω帶入得

2
1 1 1
G ( jω ) = = ⇒ ω = ωτ =
1 + ω 2τ 2 2 τ
1 1
因此我們可以得到其頻寬 B.W.= (即 0~ 間)
τ τ

c)
ω n2
1o 二階系統 G (s ) = ，根據頻寬(B.W.)定義，我們必須求得響應大
s 2 + 2ξω n s + ω n2

於 1 之範圍，推導如下：
2
 1 1 1 ω
G ( jω ) = = = ，其中 r =

1 − ω 2
2 
2

 + 4ξ 2 ω 2
2 (1 − r ) + (2ξr )
2 2 2 2 ωn
 ω  ωn
 n 
( ) (
⇒ r 4 − r 2 2 − 4ξ 2 + 1 + 8ξ 2 = 0 )
2
ω 
⇒ r =  h 
2

ωn 
= (1 − 2ξ 2 ) ± 4ξ 4 − 4ξ 2 + 2
⇒ 因為 r > 0

[
⇒ B.W . = ω n (1 − 2ξ 2 ) + 4ξ 4 − 4ξ 2 + 2 ]1
2

2o 在ξ<0.7 的條件下，我們可經由實驗數據，從 G ( jω ) MAX 求得ξ，並從共振頻率

ω r 找出 ω n 的值；再代回上面所求出之頻寬(B.W.)公式即可求得二階系統之頻寬。
1
G ( jω ) MAX =
2ξ 1 − ξ 2
ω r = ω n (1 − 2ξ 2 ) (ξ < 1/ 2 )

3. A temperature sensor can be represented as a first order element with time

constant 5 sec and steady state sensitivity 40x10-6 V/°C, calculate the output of
this sensor for the input signal given by
∆TT(t) = 75[sin (t) +1/3sin(3t)+1/5sin(5t)+1/7sin(7t)] 15%
Ans：
1o
由題目知，此一階系統具有以下之轉移函數，
~ ~
∆ O( s) ∆O ∆ T ( s )
=
∆ T T ( s ) ∆T ∆ T T ( s )
~ ~

1
= 40 × 10 −6
1 + 5s
~ 1 ~
⇒ ∆ O( s ) = 40 × 10 −6 ∆ T T ( s)
1 + 5s
3o
1
令 KG(s)= 40 × 10 −6
1 + 5s
 ~ Λ Λ
∆ T T ( s ) = I sin ω t ∆ O ( t ) = O sin( ω t + φ )
KG ( s )

Λ Λ 1 Λ
O = K G ( jω ) I = 40 × 10 −6 I
其中 1 + 25ω 2
φ = arg G ( jω ) = − tan −1 (5ω )
⇒
1
w = 1 , G(j) = = 0.196, arg G ( j ) = − tan −1 5 = −78.7 o
1 + 25
1
w = 3 , G(3j) = = 0.067, arg G (3 j ) = − tan −1 15 = −86.2 o
1 + 25 × 9
1
w = 5 , G(5j) = = 0.040, arg G (5 j ) = − tan −1 25 = −87.7 o
1 + 25 × 25
1
w = 7 , G(7j) = = 0.029, arg G (7 j ) = − tan −1 35 = −88.4 o
1 + 25 × 49
利用其線性關係，我們可以得到
1 1
∆O(t ) = 75× 40 ×10−6 × [0.196sin(t − 78.7o ) + × 0.067sin(3t − 86.2o ) + × 0.04sin(5t − 87.7o )
3 5
1
+ × 0.029sin(7t − 88.4o )]
7
−3
= 3 ×10 [0.196sin(t − 78.7o ) + 0.022sin(3t − 86.2o ) + 0.008sin(5t − 87.7 o ) + 0.004sin(7t − 88.4o )](V )

4. Suppose we use a copper-constantan thermocouple to measure small fluctuations

in temperature around 200°C. The thermocouple has a time constant of 5 sec.
The steady state sensitivity of the sensor at 200°C is 50 µV/°C
A. Find the general expression of the input-output relation of this sensor. 5%
B. Suppose the sensor has constant steady state sensitivity up to 200°C and
the equilibrium condition has been obtained at 200°C. Now the temperature is
suddenly increase from 200°C to 210°C; find an expression of the output voltage
as a function of time. 10%
Ans：
a)

dVout
T = 200 o C
= 50 µ V o C −1
dT
if τ = 5sec
 ~
∆ V out (s) 1
= 50µ
~
∆ T F (s) 1 + 5s
~ 1 ~
⇒ ∆ V out (s) = 50µ ∆ T F (s)
1 + 5s
b)
1o
Q ∆TF (t) is suddendy increasing 10 0 C
10
∴ ∆TF (s) =
s
50µ 10
⇒ ∆Vout (s) = ×
(1 + 5s) s
− 50 10
= 50 × 10 -6 ( + )
1 + 5s s
−t
⇒ ∆Vout (t) = 5 × 10 −4 (1 − e 5
)(V)

2o
Vout (t) = V(0- ) + ∆Vout (t)
−t
= 200 o C × 50V o C −1 + 5 × 10 − 4 (1 − e 5
)(V)
−t
= 0.01 + 5 × 10 − 4 (1 − e 5
)(V)
−t
= 0.0105 − 5 × 10 − 4 e 5
(V)

5. A. What is the single parameter to describe the dynamic characteristics of a first

order element? 5%
B. Give at least two methods to determine this parameter by experiments.
10%
Ans：
a)
τ(Time Constant，時間常數)
b)
1o
t = τ ,τ,f 0(t) = 0.632 ，亦即只要找出大小為 0.632 之時間點，那時的數值便是時間
常數τ的大小。
 f o (t)

0.683
response curve from experiment data

τ
2o
1
The transfer function of first-order system is G ( s ) = , and τ is time constant,
1 + τs
so we can find step response
−t
y(t) = 1 − e τ
1 −t
⇒ y' (t) = e τ
τ
1
⇒ y' (0) = for t = 0
τ
tangent curve at t=0
f o (t)

response curve from experiment data

τ
6. A force sensor has a mass of 0.5kg, stiffness of 2x102 Nm-1 and a damping
constant of 6.0N s m-1.
A. Calculate the steady-state sensitivity, natural frequency and damping ratio for
the sensor. 5%
 B. Calculate the displacement of the sensor for a steady input force of 5N. 5%
C. If the input force is suddenly increased from 5 to 10 N, derive an expression
for the resulting displacement of the sensor. 5%
D. From C, what about if after 10sec, the force is suddenly decreased from 10N
back to 5N? Derive an expression for the resulting displacement of the
sensor in this situation. 5%
Ans：
a)
由題目知 m= 0.5(kg)、K=2x102 (Nm-1 )、λ= 6.0(N s m-1)，帶入公式可得：
k 200
natural frequency ω n = = = 20( rad )
m 0.5 sec
λ 6.0
damping ratio ξ = = = 0.3
2 km 2 200 × 0.5
1 1
steady - state sensitivit y K = = = 5 × 10 −3 (mN −1 )
k 200

b)
displacement = F × K = 5 × 5 × 10 −3 = 0.025(m) = 2.5(cm)

c)
Because of ξ = 0.3 < 1 , so that from eq 4.33 and eq 4.31, we can find that：

⇒ ∆x(t) = steady - state sensitivity × step height × unit step response f 0 (t)
⇒ displacement = steady − state displacement + ∆x(t)
ξ
= 2.5 + 5 × 10 −3 × 5 × 100 × {1 − e −ξω n t [cosω n 1 − ξ 2 t + sinω n 1 − ξ 2 t]}
1 −ξ 2

0.3
= 2.5 + 5 × 10 −3 × 5 × 100 × {1 − e −6t [cos20 1 − 0.3 2 t + sin20 1 − 0.3 2 t]}
1 − 0.3 2

− 6t
= 2.5 + 2.5{1 − e [cos19t + 0.32sin19t]}(cm)

d)
1o
After 10seconds，steady-state displacement become 5cm。

2o
displacement = steady − state displacement + ∆x(t)
ξ
= 5 - 5 × 10 −3 × 5 × 100 × {1 − e ξω n t [cosω n 1 − ξ 2 t + sinω n 1 − ξ 2 t]}
1 −ξ 2

0.3
= 5 − 5 × 10 −3 × 5 × 100 × {1 − e − 6t [cos20 1 − 0.3 2 t + sin20 1 − 0.3 2 t]}
1 − 0.3 2

− 6t
= 5 − 2.5{1 − e [cos19t + 0.32sin19t]}(cm)