1

Analogue signals acquisition

My slides are subject to change without prior notice

References

• U. Tietze, Ch.Schenk, Electronics Circuits – Handbook for Design and Applications, Springer, 2010

• Advertisement materials and Application notes of:

– Linear Technology,

– Analog Device,

– Texas Instruments,

– National Semiconductors

– NEV Corporation

Outline

• Aquisition system –general diagram

• ADC – general information

• Isolation methods

• Multiplexing problems

• Sample & hold – sampling frequency

• Widowing

2

Intoduction

The intent of this and a few next lectures is to introduce the students to the most commonly used transducer interfaces and to provide practical information for dealing with the most frequently encountered transducers and their associated signals.

Transducers are available for measuring many physical quantities, such as temperature, pressure, strain, vibration, sound, humidity, flow, level, velocity, charge, pH, and chemical composition, among others. In most cases, the transducer manufacturer provides application notes on the transducer’s use and principles of operation. The main questions to consider when selecting a transducer are:

• the electrical characteristics - source impedance, amplitude, frequency,

• power supply or excitation

• transducer’s accuracy (taking frequency and range into account)

•calibration method

• verification of calibration

• environment of operation (temperature, humidity, vibration, pressure, etc.)

General controll schematic outline

OBJECT

UNDER

CONTROLL

Phisical

enviroment

Front-end

circuit

•impedance transformer

•amplifier

•multiplexer

•isolation

ADC computer

PID

controlleractuator DAC

This elements can have different configuration

!!!!

analogue

isolationdigital

isolation

ADC Analog to Digital Converters

• parallel converter – flash

• successive approximation

• pipeline

• voltage to frequency

• integrating converter (two, four slop)

• delta-sigma converter

3

ADC – parallel (flash)

2-step converter(direct, serial, pipeline)

Features (compared to „flash”):

•lower number comparators

•higher conversion time

•more popular (cheaper)

•higher resolution

3-step converter

(direct, serial, pipeline)

4

ADC - successive approximation

ADC - voltage to frequency

frequency meter

ADC - integrating converter

5

Indirect converter ---four - slope

Phase 0 1 2 3 4

UR GND UR Ui UR

Four slope converter - „principle” of corrections

measured voltage

numerical result

REF

REF

REF

REF

UoffsetUresult

offsetresult

offsetresultoffsetUresult

Uout

444 3444 21

4444444 34444444 21

44 344 21

321

1.......0

)3(

)2(

)1(

)(

)(

−−=

=

−

−=

(1)

(2)

(3)

offset

result

(UREF)

(UREF)

„Σ-Δ” converter

oversampling converter

VAD

VDA

!!!!!!!!

Sampling frequency fOSR is much greater

than reading frequency fS

for example

44,29 MHz

for example 44,1

kHz 16 bit

6

„Σ-Δ” converter

- principle of operationfOSR

Vi

VDA

VAD

Oversampling „Σ-Δ” converter

- general principle

Samplig

frequencyResolution

fOSR N

fS<<fOSR N+x

Noise shaping in „Σ-Δ” converter

7

„Σ-Δ” converter – linear model

Conventional converter

S

OS

Nout

ff

U1∝

bitdB

ff

NS

OS

bits/6

log20

=fOS/fS bit

2 0.5

4 1

256 4

First order „Δ-Σ” converter3

1

∝

S

OS

Nout

ff

U

bitdB

ff

NS

OS

bits/6

log20

3

=

fOS/fS bit

2 1.5

4 3

256 12

8

3dr order „Δ-Σ” converter5

1

∝

S

OS

Nout

ff

U

bitdB

ff

NS

OS

bits/6

log20

5

=

fOS/fS bit

2 2.5

4 5

256 20

Oversampling

oversampling ratio

number od bits

(resolution) 1st order „Δ-Σ” converter

traditional

converter

2nd order „Δ-Σ” converter

Noise shaping in „Δ-Σ” converter

9

Example of „Δ-Σ” converter

ADS1110

Lab setup of pressure sensor

General comparison of

AD converters

sampling rate

resolution (bits)

24

20

16

12

8

4

0

1 1k 1M 1G

dual slop

counting

Δ-Σ

successive

approximation

pipeline

flash

10

Resolution

UFS

Δ

ADC – resolution and SNR

12

2

12

1)(

2

22

2

22

=∆=∆

== ∆

∆−

∞

∞−

nFS

N

U

dxxdxxpxP

-Δ/2 Δ/2

1/Δ

][76,102,6

12

2

2

2

log10log102

2

dBnU

U

P

PSNR

nFS

FS

N

S +⋅=

==

ADC ----SNR

][76,102,6

12

2

2

2

log10log102

2

dBnU

U

P

PSNR

nFS

FS

N

S +⋅=

==

][02,62log20 dBSNR ==For every bit of resolution SNR increases by 6dB !!!!!!!!

11

Channel Capacity [bit / second]

3)2(log10)2(log20

log20

2)2(log

log

2

log2sec]/[

1010

10

10

10

2

SNRBSNRB

U

U

BU

U

B

U

UBnfbitsCC

ampamp

amp

bitssampl

⋅==

∆=

∆=

=

∆==

sec]/[3

bitsSNRB

CC⋅=

Shannon–Hartley theorem

ADC Errors

• offset

• linearity

• gain

• missing codes

• noise (interferences)

12

ADC Errors - offset

Voltage

Reading

Uoff

ADC Errors – nonlinearity

Voltage

Reading

ADC Errors – gain

Voltage

Reading

13

ADC Errors – missing codes

Voltage

Reading

ADC Errors - noise (interferences)

Controlling noise is first step to measure anything.

Spending money for accurate and high resolution

ADC can be worthless when measuring signal is

interfered by high level noise.

noise level should be smaller than resolution

Digital to analog conversion

39

14

Digital to Analogue Conversion

the parallel method the weighting method computing method

Weighting conversion

1+−=

max

refoutN

NUU TIP:

difficult for manufacturing because of different R;

Uref is loaded depending N;

stray capacitance of switches are charged or dis-

every change;

Weighting conversion

TIP:

difficult for manufacturing because of different R;

Uref is loaded depending N;

stray capacitance of switches are charged or dis-

every change;

15

Ladder network

TIP:

difficult for manufacturing because of different R;

Uref is loaded depending N;

strai capacitance of switches are charged or dis-

every change; 1+−=

max

refoutN

NUU

Ladder network

converter without any amp

Isolation

• Optical

• Magnetic

• Capacitive

• Analogue

• Digital

16

„Digitaly” isolatet analog signals

Opto-Isolation of analog signals

Analog opto-isloation

1

1R

UI in

PD =2

2R

UI out

PD =12 PDPD II η=

η1

2

R

RUU inout = Tip:

accuracy can be is not

satisfactory

17

Opto-isolatin of 4-20mA loop

PWM to 4-20mA isolation

MAXIMy

advertisement materials

Magneto - Isolation

NVE corporation

Advertisement materials

18

Isolation by Pulse With Modulation (PWM) (Burr Brown – advertisement materials)

Isolation by modulation(Analog Device – advertisement materials)

Voltage-frequency modulation in

transmission and isolation

isolator

U/f

Procesor

T/f aquisition

long

transmission

line

fibre

ISOLATION

!!!!!

19

Multiplexing

• direct (small signal) multiplexing

• buffering – multiplexing

• simultaneous s/h

• digital multiplexing

direct multiplexingsingle ended

or differential

inputs

MUX

GAIN !!

different for

different signals

ADDRES

ADC

Input buffering

MUX

GAIN !!

not nessesery

!!!

ADDRES

ADC

single ended

or differential

inputs

singe ended

inputs

GAIN !!

different for

different signals

20

multiplexing with simultaneous

reading

MUX

ADDRES

ADC

single ended

or differential

inputs

S/H

S/H

S/H

SAMPLING

Switches to be continued during next lectures

UCON=VCC

UCOFF=0

RON<100Ω

SH – sample and hold – basic circuitto be continued during next lectures

60

Rout=0 Rin=∞

21

Time acquisition channel

ADC

external settle time T1 [ns-us] Settle time of ADC T2 [ns-us]

MUX

settlesample TTTT =+≥ 2

2

2

1samplechannel

acqTN

f⋅

= 1

Summary

• Aquisition system –general diagram

• ADC – general information

• isolation methods

• Multiplexing problems

• Sample & hold – sampling frequency

Test questions (an example):

• In digital to analog conversion, what is the

signal to quantization noise ratio ?

• What is the relationship between SNR and

Channel Capacity ?

• Describe relationship between sampling ratio

and resolution of AD converters.

• What are basic principles (methods) of galvanic

isolation in analog and digital systems ?

22

Multiplexers & switches –

more about in

„AD/DA converters lecture”

Windowing

(facultative)High- and moderate-resolution windows

1 Rectangular window

2 Hann window

3 Hamming window

4 Tukey window

5 Cosine window

6 Lanczos window

7 Triangular windows

8 Gaussian windows

9 Bartlett–Hann window

10 Blackman windows

11 Kaiser windows

Windowing

Low-resolution (high-dynamic-range) windows

1 Nuttall window, continuous first derivative

2 Blackman–Harris window

3 Blackman–Nuttall window

4 Flat top window

23

Windowing

Other windows

1 Bessel window

2 Dolph-Chebyshev window

3 Hann-Poisson window

4 Exponential or Poisson window

5 Rife-Vincent window

6 DPSS or Slepian window

Main and side lobes

Windowing

rectangular window

24

Windowing

Hann window

Windowing

Hamming window

Stopband attenuation of different windows