Transcript Asics for MEMS

Asics for MEMS
BRILLANT Grégory
2th of October 2006
Overview
I.
Smart Sensor Interface
Electronics
II.
Equivalent circuit representation of
electromachanical transducers
October 2006
Smart Sensor Interface
Electronics
Overview
I.
II.
III.
IV.
V.
October 2006
Object Oriented Design
Parasitic Effects
Analog to Digital Conversion
High accuracy over a wide
Dynamic range
Presentation of two systems
Introduction
Information processing systems need sensors to
acquire physical, mechanical and chemical
information
Sensors are inescapable in applications such as
smart cars or smart homes
But: They areEXPENSIVE
Solution: Smart Sensor Systems combine :
Sensors
Signal conditioning
ADC
Bus interfacing
And self testing, auto-calibration, data evaluation and
identification, …
October 2006
Object oriented Design
of sensor system
When designing sensor systems, traditional
Top/Down and Bottom/Up approach are limited
Interdisciplinary and open characters of sensor
subsystems
Long design time and inflexible designs
Solution: Object oriented Design.
The result of the object-oriented design is a detail
description how the system can be built, using objects
Save costs and speed up the design
If it’s possible to implement the sensor element
and its interface on a same chip, we speak about
Smart Sensor
October 2006
Parasitic effects in
sensing elements
Excitation signals for sensing elements are
usually square-wave (and not sinusoidal).
Care has to be taken to avoid undesired
electro-physical interaction
Electrical excitation of a resistive temperature
sensor causes self heating → measurement
errors
In conductivity sensors the excitation can
cause electrolysis → corrosion
October 2006
Parasitic effects in
sensing elements
Sensing elements deliver an electrical output
representing the measurement
But: there are parasitic electrical effects
Capacitive humidity sensors are often shunted by a
parasitic resistive component
Resistive sensors are often shunted by parasitic
capacitors
The various components are founded by
analyzing impedance measurement at various
frequencies.
Small-sized, low power integrated circuit must be
used to achieve this measurements
The use of additional sensor elements improve
the reliability
October 2006
Parasitic effects in
sensing elements
Connecting wires and cables can affect the
measurement
Solution: two-port measurements
4-wire technique is applied to
measure a low-ohmic sensor.
The interface chip delivers an
excitation current and the
voltage over the sensor is
measured using a high
impedance input amplifier
The dual case is applied to
measure a high-ohmic sensor
admittance
October 2006
Analog to digital conversion
The sensor signal is often converted to a
voltage signal → Standard ADC can be
used
Capacitive sensing element: A/D converter
requires an analog input voltage →
Problem: Complication of the front end ADC
design because of the introduction of:
Many additional transfer parameters
Biasing quantities
Conversion steps
October 2006
Analog to digital conversion
Solution: sample and hold, quantization, digital
filtering and ∑Δ conversion can be implemented
in the DSP microcontroller
DSP or microcontroller are well equipped to
measure frequency or time interval → using of
period-modulated signal
October 2006
High accuracy over a wide
Dynamic range: errors
Two kinds of errors:
Systematic errors: inaccuracy of the system
parameters → calibrating
Random errors: interferences, noise and
instability → filtering, separating common mode and
differential mode signal,…
Calibration: sensor-under-test is compared to
another one of superior quality
Trimming: sensor behavior is altered
permanently to make its characteristics match
the nominal as close as possible
October 2006
High accuracy over a wide
Dynamic range: chopping
But: calibration and trimming have to
performed under certain conditions with
respect to the temperature, supply voltage
and time conditions during sensor
operation
Solution:
Chopping +,-,+,-,…
techniques. Reduce
Common chopper:
random
errors,
noise,
low frequency
Improved
chopper:
+,-,-,+,+,…
interferences and offset
This sampling sequence result in a filter
Implementation: switches interchange the
operation
to the at
interferences
wires
of a applied
signal source
a high
≠
frequency
October 2006
High accuracy over a wide
Dynamic range: auto calibration
Two signal approach: measure of a
reference signal S1 in exactly the same
way as the input signal Sx
Ratio Sx/S1 or difference Sx-S1 is used
to eliminates errors
Three signal approach: more accurate.
Measure of two reference signals.
(Sx-S1)/(S2-S1) is used.
October 2006
High accuracy over a wide
Dynamic range: amplification
During auto-calibration, the signals are
processed in an identical way
The system should be linear or well
characterized over the full signal range →
this poses a problem when the signal are not in
the same range of magnitude
To achieve a high accuracy, signals should have
a high dynamic range, but that is not often the
case
Amplification or division by a scaling factor A
October 2006
High accuracy over a wide
Dynamic range: amplification
Problem: realize A without loosing
precision
Dynamic feedback instrumentation
amplifier can solve this problem
Resistive load K=u+v+w+z
Dynamic feed back is made by rotation of
the resistor chain
Mismatches between resistors are critical
6*K switches
October 2006
High accuracy over a wide
Dynamic range: amplification
DEM amplifier can also be a solution
Possible implementation: switchedcapacitors
The rotation of the capacitors at each
clock cycle can almost eliminates the
effect of capacitor mistmatching
October 2006
High accuracy over a wide
Dynamic range: division
Instead amplify the smallest signals,
division of the strongest signals can also
be applied
One possible realization
A resistive voltage divider (Nr resistors)
combined with a capacitive voltage divider (Nc
capacitors)
Division ratio: NcNr
October 2006
Universal sensor interface
The Universal Sensor Interface is a
complete front-end for sensor systems
The output is based on a period modulator
oscillator
The USI converts the signals of sensing
elements into period-modulated signals →
microcontroller and DSP compatible
Signal processing in the USI
The input signal is selected by the multiplexer
Chopped signal conversion
Period length conversion
October 2006
A wide range voltage processor
Example: measurement system for
thermocouple voltages
Two measured signals: thermocouple
voltage Vx and reference junction
temperature Tj
Voff is measured to allow offset
compensation
All algorithmic signal processing is
performed by the microcontroller. The
voltages are firstly converted to the time
domain
October 2006
Conclusion
In the smart sensor systems presented,
measurement techniques are implemented
using a limited number of low-cost, lowpower integrated circuits only.
By applying synchronous detection, auto
calibration and advanced chopping, high
immunity is obtained for interfering
signals, 1/f noise and parameter drift.
The dynamic range of the signals can be
extended using dynamic amplifiers and
dynamic dividers.
October 2006
Equivalent circuit
representation of
electromachanical
transducers
Overview
I.
II.
III.
IV.
V.
October 2006
Lumped-parameter
electromechanical systems
Elementary Lumped-parameter
transducers
Equivalent circuit
representation
Coupling of the transducers
to the outside world
Some examples of transducers
Introduction
A transducer is a device that converts one type of
energy to another, or responds to a physical
parameter. A transducer is in its fundamental
form a passive component.
Electomechanical transducers are used to convert
electrical energy into mechanical energy and vice
–versa
Example: microphone in which a sound pressure
is converted into an electrical signal
Equivalent circuit approach: the electrical and
mechanical portions of the transducers are
represented by electrical equivalents → single
representation of device that operate in more than one
energy domain.
October 2006
Lumped-parameter
electromechanical systems
Lumped parameter (or discrete) system:
physical properties (mass, capacitance,
inductance,…) are concentrated or lumped
into single physical elements
The parameters which involve ordinary
differential equations are called linear
lumped parameters.
Lumped-parameter modeling is valid as
long as the wavelength of the signal is
greater than all dimensions of the system
Example: basic configuration of an
electrostatic transducer
October 2006
Energy exchange
Exchange of energy of a transducer and
the outside world is achieved trough
ports: pair of conjugate dynamic
variables, the effort variable and the flow
October 2006
Elementary Lumped-parameter
transducers: configurations
Linear transducers are mathematically
more easiest to study
Linear behavior is achieved for small
signal variations around equilibrium levels
Four basics electromechanical lumpedparameter transducers:
Transverse electrostatic
transducer
In-plane electrostatic
transducer
Electromagnetic transducer
Electrodynamics transducer
October 2006
Elementary Lumped-parameter
transducers: equations
Characteristic equations: linear relations
between small-signal variations of the port
variable around the bias point
Matrix representations
Matrix B: effort variable as a function of
state variable
Matrix T: relates the effort-flow variables
at the electrical port directly to those at
the mechanical port
October 2006
Elementary Lumped-parameter
transducers: equations
The coupling factor k represents the
electromechanical energy conversion in
lossless transducers
A coupling factor of 0 means no
interactions
A state of equilibrium exists for 0<k<1
Typical values for k are between 0.05 and
0.25
October 2006
Equivalent circuit
representation
There is an analogy in the mathematical
descriptions between electric and mathematical
phenomena
A series connection in the mechanical circuit
becomes parallel in the equivalent electrical
circuit
October 2006
Equivalent circuit
representation
The construction of the equivalent networks
starts with the splinted transfer matrix of the
electrostatic transducers
Center matrix: transducer
Left matrix: electrical admittance
Right matrix: mechanical impedance
Each of the constituent can be represented by an
equivalent network
October 2006
Equivalent circuit
representation
There is no one way to decompose a matrix
But: each decomposition has its own network
representation
The choice of which circuit to use is dictated by
the application
October 2006
Coupling of the transducers
to the outside world
The exchange of energy of the transducer and
the outside world is done trough ports
Laws of equilibrium link the transducer via their
port relation to the external elements
Electrical parts: Kirchoff’s voltage and current
laws
Mechanical parts: Newton law (∑Fi=0) and
continuity of space (∑Ui=0, U: incremental
velocity)
The mechanical laws are directly obtained by
invoking the Kirchoff’s laws to the mechanical
part in the equivalent circuit representation
October 2006
Examples of lumped-parameters
electromecanical sytems
I.
II.
III.
IV.
October 2006
Condenser Microphone
In-plane parallel microresonators
Vibration sensors
Electromechanical feedback
Condenser microphone,
force and displacement transducers
Operating principles:
the force to be measured is exerted on the mass.
The motion of the mass is converted into an electrical signal, a
current, which flows in part through a resistors → production
of an output voltage.
This voltage is a measure for the applied force or
displacement.
October 2006
Condenser microphone,
force and displacement transducers
October 2006
Condenser microphone,
force and displacement transducers
If the applied force is the result of an
acoustic pressure, the transducer can be
used as an electrostatic or condenser
microphone
October 2006
In-plane parallel
microresonators
In-plane parallel
microresonators using
electrostatic
interdigitated
structures for
excitation and
detection of the
vibrational motion are
used as transducing
elements in a wide
variety of applications
October 2006
Vibration sensors
Vibration sensors are
employed for
measurements on moving
vehicles, on buildings, or
on machinery or as seismic
pickups
The basic principle of
vibration measurements is
simply to measure the
relative displacement of a
mass connected to the
vibrating body.
The transducer detects the
mass displacement Xm
relative to the
displacement Xin of the
vibrating body
October 2006
Electromechanical feedback
Electromechanical
feedback (or force
balancing) is often
employed for
applications requiring
a great accuracy
The system measures
the force Fm exerted
directly on the mass
The upper capacitor
senses the induced
mass displacement
resulting in a change
of the plate charge
October 2006
Electromechanical feedback
The output voltage of the
charge amplifier Va is next
amplified by a high-gain
(servo) amplifier T
The output voltage Vout is
fed back to the lower
capacitor.
This generates an
electrostatic force which is
proportional to the relative
mass displacement and
which always opposes
motion of the mass from
the rest position.
This way, the mass itself is
kept very close to the
zero-displacement position
October 2006
Conclusion
The majority if the circuits presents 3
different parts:
An electrical part
An electromechanical coupling part
A mechanical part
The equivalent circuits can be used to
determine the frequency and the transient
response of the transducer
The equivalent circuit theory applied to
the study of transducer characteristics is a
basis for further investigations
October 2006