Experience Seamless Full Ebook Downloads for Every Genre at textbookfull.

com

Mixed Design of Integrated Circuits and Systems

The Springer International Series in Engineering
and Computer Science 434 1998th Edition Andrzej
Napieralski Zygmunt Ciota Augustin Martinez
Gilbert De Mey Joan Cabestany
https://textbookfull.com/product/mixed-design-of-integrated-
circuits-and-systems-the-springer-international-series-in-
engineering-and-computer-science-434-1998th-edition-andrzej-
napieralski-zygmunt-ciota-augustin-martinez-gilbert-de/

OR CLICK BUTTON

DOWNLOAD NOW

Explore and download more ebook at https://textbookfull.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Advanced Low Power Digital Circuit Techniques The Springer

International Series in Engineering and Computer Science
405 Elrabaa Muhammad S Abu Khater Issam S Elmasry Mohamed
Ihttps://textbookfull.com/product/advanced-low-power-digital-circuit-
techniques-the-springer-international-series-in-engineering-and-
computer-science-405-elrabaa-muhammad-s-abu-khater-issam-s-elmasry-
mohamed-i/
textboxfull.com

Ambient Communications and Computer Systems Gregorio

Martinez Perez

https://textbookfull.com/product/ambient-communications-and-computer-
systems-gregorio-martinez-perez/

textboxfull.com

Human Language Technology Challenges for Computer Science

and Linguistics Zygmunt Vetulani

https://textbookfull.com/product/human-language-technology-challenges-
for-computer-science-and-linguistics-zygmunt-vetulani/

textboxfull.com

Academic English for Computer Science An English for

Specific and Academic Purposes Course for International
students of Computer Science Computer Engineering
Information and Communication Systems Rizopoulou
https://textbookfull.com/product/academic-english-for-computer-
science-an-english-for-specific-and-academic-purposes-course-for-
international-students-of-computer-science-computer-engineering-
information-and-communication-systems-riz/
textboxfull.com
Enabling the Internet of Things From Integrated Circuits
to Integrated Systems 1st Edition Massimo Alioto (Eds.)

https://textbookfull.com/product/enabling-the-internet-of-things-from-
integrated-circuits-to-integrated-systems-1st-edition-massimo-alioto-
eds/
textboxfull.com

Design of CMOS Analog Integrated Fractional Order Circuits

Applications in Medicine and Biology 1st Edition Georgia
Tsirimokou
https://textbookfull.com/product/design-of-cmos-analog-integrated-
fractional-order-circuits-applications-in-medicine-and-biology-1st-
edition-georgia-tsirimokou/
textboxfull.com

Design of Analog CMOS Integrated Circuits 2nd eddition 2nd

Edition Behzad Razavi

https://textbookfull.com/product/design-of-analog-cmos-integrated-
circuits-2nd-eddition-2nd-edition-behzad-razavi/

textboxfull.com

Durability of Composite Systems (Woodhead Publishing

Series in Composites Science and Engineering) 1st Edition
Kenneth Reifsnider (Editor)
https://textbookfull.com/product/durability-of-composite-systems-
woodhead-publishing-series-in-composites-science-and-engineering-1st-
edition-kenneth-reifsnider-editor/
textboxfull.com

Advances and Applications in Computer Science, Electronics

and Industrial Engineering: Proceedings of the Conference
on Computer Science, Electronics and Industrial
Engineering (CSEI 2019) Jyrki Nummenmaa
https://textbookfull.com/product/advances-and-applications-in-
computer-science-electronics-and-industrial-engineering-proceedings-
of-the-conference-on-computer-science-electronics-and-industrial-
engineering-csei-2019-jyrki-numme/
textboxfull.com
MIXED DESIGN OF INTEGRATED
CIRCUITS AND SYSTEMS
 THE KLUWER INTERNATIONAL SERIES
IN ENGINEERING AND COMPUTER SCIENCE
MIXED DESIGN OF INTEGRATED
CIRCUITS AND SYSTEMS

edited by

Andnej Napieralski
Zygmunt Ciota
Technical University of LOdi
LOdi, Poland

Augustin Martinez
Laboratoire d 'Analyse et d 'Architecture des Systemes
du Centre National de la Recherche Scientifique
Toulouse, France

Gilbert De Mey
University of Gem
Gent, Belgium

Joan Cabestany
Universitat Po/itecnica de Catalunya
Barcelona, Spain

SPRINGER SCIENCE+BUSINESS MEDIA, LLC

ISBN 978-1-4613-7586-9 ISBN 978-1-4615-5651-0 (eBook)
DOI 10.1007/978-1-4615-5651-0

Library of Congress Cataloging-in-Publication Data

A C.I.P. Catalogue record for this book is available

from the Library of Congress.

Copyright © 1998 by Springer Science+Business Media New York

Originally published by Kluwer Academic Publishers in 1998

AII rights reserved. No part of this publication may be reproduced, stored in a

retrieval system or transmitted in any form or by any means, mechanical, photo-
copying, recording, or otherwise, without the prior written permission of the
publisher, Springer Science+Business Media, LLC.

Printed on acid-free paper.

 CONTENTS

I. ANALOG CIRCUITS DESIGN

1. MULTICHANNEL Low NOISE, Low POWER ANALOGUE READOUT CHIP

FOR SILICON STRIP DETECTORS................................................................................ 3
WOJCIECH BIALAS, WLADYSLAW DABROWSKI, PAWEL GRYBOS AND MAREK IDZIK

2. DESIGN METHODOLOGY FOR CURRENT CONVEYOR BASED

CONTINUOUS-TIME FIELD-PROGRAMMABLE ANALOG ARRAy ................................ 9
RICHARD GRISEL, CHRISTOPHE PREMONT, NACER MOUCHI AND JEAN-PIERRE CHANTE

3. RF AMP IC FOR OPTICAL DISC PLAYER ................................................................ 15

CHUN-SUP KIM, GEA-OK CHO, YONG-HWAN KIM AND BANG-SUP SONG

4. CMOS CURRENT CONVEYOR DESIGN AND MACROMODEL. .................................. 21

STANISLAW KUTA, WITOLD MACHOWSKI AND ROBERT WYDMANSKI

5. BEHAVIOURAL NOISE MODELLING OF CROSS-COUPLED RING OSCILLATORS ••••••• 29

LESZEK 1. OPALSKI

6. A 27MHZ FULLy-BALANCED OTA-C FILTER IN 2/lM CMOS TECHNOLOGY •••••••• 35

BOGDAN P ANKIEWICZ, JACEK JAKUSZ AND STANISLAW SZCZEPANSKI

7. PROGRAMMING ANALOG NON-VOLATILE MEMORIES •••••••••••••••••••••••••••••••••••••••••• .41

ERIC TOURNIER AND JEAN-LOUIS NOULLET

8. FOUR-QUADRANT CMOS AMPLIFIER FOR LOW-VOLTAGE

CURRENT-MoDE ANALOG SIGNAL PROCESSING .................................................... 47
RYSZARD WOJTYNA, PIOTR GRAD AND JAROSLAW MAJEWSKI

II. POWER DEVICES AND THERMAL ASPECTS

9. ApPLICATION OF INVERSE PROBLEMS TO IC

TEMPERATURE ESTIMATION ................................................................................... 55
MARCIN JANICKI, MARIusz ZUBERT, WOJCIECH W6JCIAK, MARIUSZ ORLIKOWSKI
AND ANDRZEJ NAPIERALSKI

10. THERMAL MODEL FOR MCM's .............................................................................. 61

FRANCESC MASANA

11. CONTRIBUTION OF RADIATION IN HEAT DISSIPATION

IN ELECTRONIC DEVICES ........................................................................................ 69
BOGUSLAW WIIlCEK AND GILBERT DE MEY

12. MODELLING AND SYNTHESIS OF ELECTRO-THERMAL MICRODEVICES ••••••••••••••••• 77

WOJCIECH W6JCIAK, ANDRZEJ NAPIERALSKI, MARIUSZ ORLIKOWSKI
AND MARIUSZ ZUBERT
vi Contents

III. MICROSYSTEMS AND NEURAL NETWORKS

13. LAYOUT OPTIMIZATION OF CMOS PHOTOTRANSISTORS •••••••••••••••••••••••••••••••••••••• 85

M. MORENO, S.A. BOTA, R. HOLGADO, A. HERMS AND 1. CALDERER

14. MUL TlLAYER PIEZOELECTRIC SENSORS ON THE BASIS

OF THE PZT TYPE CERAMICS ................................................................................ 91
DIONIZY CZEKAJ, JULIAN DUDEK, ZYGMUNT SUROWIAK, ALEKSANDR V. GORISH,
YURIN. KOPTEV, ALEKSANDRA. KUPRIENKO, ANATOLI E. PANICH

15. ARTIFICIAL NEURAL NETWORK MIXED-SIGNAL PROTOTYPE SYSTEM

FOR MODEL PARAMETER IDENTIFICATION ............................................................. 97
ANDRZEJ MATERKA, PAWEL PELCZYNSKI AND MICHAL STRZELECKI

16. MIXED AID VLSI ARCHITECTURE FOR THE EMULATION

OF NEURO-FUZZY MODELS ................................................................................... 103
JUAN MANUEL MORENO, JORDI MADRENAS, SPARTACUS GoMARIZ
AND JOAN CABESTANY

17. A NEW ApPROACH FOR FINDING OPTIMUM DESIGN OF ELECTROSTATIC

MICROMOTORS •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 109
ABDUL WAHAB A. SALMAN, ANDRZEJ NAPIERALSKI, MAREK TuROWSKI
AND GRZEGORZ JABLONSKI

18. ONE-CYCLE CONTROLLED BOOST CONVERTER FOR MICROSYSTEMS ••••••••••••••••• 115

NOUREDDINE SENOUCI, FRANCIS THEREZ AND DANIEL ESTEVE

IV. DESIGN METHODOLOGIES

19. DESIGN FOR REUSE: HDL BASED GRAPHIC DESIGN ENTRY FOR
PARAMETRIZABLE AND CONFIGURABLE MODULES (A CASE STUDY) •••••••••••••••••••• 125
ANNA BOSZKO

20. HIERARCHICAL TEST GENERATION FOR DIGITAL SYSTEMS••••••••••••••••••••••••••••••••• 131

MARINA BRIK, GERT JERVAN, ANTII MARKUS, JAAN RAIK AND RAIMUND UBAR

21. PATH SELECTION BASED ON INCREMENTAL TECHNIQUE ••••••••••••••••••••••••••••••••••••• 137

S. CREMOUX, N. AzEMARD AND D. AUVERGNE

22. CHIP AREA ESTIMATION FOR SC FIR FILTER STRUCTURES

IN CMOS TECHNOLOGY ....................................................................................... 143
ADAM DJ\BROWSKI AND RAFAL DLUGOSZ

23. SIMPLIFIED MODELS OF IC's FOR THE ACCELERATION OF CIRCUIT DESIGN ••••• 149
VLADIMIRA. KOVAL, MYKOLAB. BLYZNIUKANDIRENA Y. KAzYMYRA

24. Low POWER METHODOLOGIES FOR GAAs ASYNCHRONOUS SYSTEMS•••••••••••••••• 157

STEFAN W. LACHOWICZ, KAMRAN ESHRAGHlAN, JOSE F. L6pEZ, ROBERTO SARMIENTO
AND HANS JORG PFLEIDERER

25. TRANSLATION OF C AND VHDL SPECIFICATIONS INTO INTERPRETED

PETRI NETS FOR HARDWARE/SOFTWARE CODESIGN ........................................... 163
JAROSLAW MIRKOWSKI AND ZBIGNIEW SKOWRONSKI
 Contents vii
26. FIPSOC. A NOVEL MIXED FPGA FOR SYSTEM PROTOTYPING .......................... 169
J.M. MORENO, J. CABESTANY, E. CANT6, J. FAURA, P. VAN DUONG, M.A. AGUIRRE
AND J.M. INSENSER

27. A NORDIC PROJECT ON HIGH SPEED Low POWER DESIGN

IN SUB-MICRON CMOS TECHNOLOGY FOR MOBILE PHONES ............................. 175
OLE OLESEN

28. A REUSE CONCEPT FOR AN I2C-BuS INTERFACE ................................................. 181

M. PADEFFKEAND W. GLAUERT

29. DYNAMIC ANALYSIS OF DIGITAL CIRCUITS WITH 5-VALUED SIMULATION ......... 187
RAiMUND UBAR

v. ADVANCED TRENDS IN MICROELECTRONICS EDUCATION

30. A FINITE STATE DESCRIPTION OF THE EARLIEST LoGICAL COMPUTER:

THE JEVONS' MACHINE ........................................................................................ 195
PAUL AMBLARO

31. EDUCATIONAL COMPUTER USE IN THE MOS TRAINING FAB .............................. 203
A. FERRElRA-NoULLET, P.F. CALMON, J.L. NOULLET

32. TEACHING OF ANALOG IC DESIGN WITH MODERN CAD TOOLS

AND CMOS PROCESSES ........................................................................................ 209
JORGEN FRICKEL, JAFAAR ME.JRI AND WOLFRAM GLAUERT

33. TEACHING POWER ELECTRONICS WITH TwO-DIMENSIONAL

SEMICONDUCTOR DEVICES MODELS ..................................................................... 215
MARIusz GRECKI, GRZEGORZ JABLONSKI, MAREK TuROWSKI
AND ANDRZEJ NAPIERALSKI

34. VLSI Top-DoWN DESIGN FOR STUDENTS OF COMPUTER SCIENCE-

A PRACTICAL COURSE .......................................................................................... 221
IGOR KATCHAN, FRANK MAYER AND DETLEF SCHMID
 Preface

Very fast advances in IC technologies have brought new challenges into the physical
design of integrated systems. The emphasis on system performance, in lately developed
applications, requires timing and power constraints to be considered at each stage of
physical design. The size of ICs is decreasing continuously, and the density of power
dissipated in the circuits is growing rapidly.
The first challenge is the Information Technology where new materials, devices,
telecommunication and multimedia facilities are developed. The second one is the
Biomedical Science and Biotechnology. The utilisation of bloodless surgery is possible
now because of wide micro-sensors and micro-actuators application. Nowadays, the
modern micro systems can be implanted directly into the human body and the medicine can
be applied right in the proper time and place in the patient body. The low-power devices
are being developed particularly for medical and space applications. This has created for
designers in all scientific domains new possibilities which must be handed down to the
future generations of designers.
In this spirit, we organised the Fourth International Workshop "MIXED DESIGN
OF INTEGRATED CIRCUITS AND SYSTEMS" in order to provide an international forum for
discussion and the exchange of information on education, teaching experiences, training
and technology transfer in the area of microelectronics and microsystems.
The objective of this Conference was to discuss all the problems which have to be taken
into account by future designers of modern integrated circuits and devices. Different
aspects of the advanced electronic design, testing and manufacturing, which should be
reflected in the high-level education, have been presented. Therefore the topics of this
Conference have been very carefully chosen. The main topics of discussion were:
• Design methodologies, techniques, and practical examples
• Thermal problems in semiconductor devices and ICs
• Analog and digital filters and signal processing
• Advanced testing methods and reliability
• Neural networks
• Sensors, actuators and micromachines: modelling, simulation and design
• Power devices and Smart-Power modules
• New trends in microelectronics: RF, Low Power, Low Voltage
• Education of CAD of modern IC devices
• EUROPRACTICE activities
All these topics are very important for the general knowledge of students. Future engineers
must be able to think not only about the common utilisation of the specific VLSI design
tools, but they have to think about all their design environment as wel\. In modern
electronics, testing problems, thermal problems, sensors and actuators design problems
and all aspects of power integration design are of great importance.
x Preface

The Conference covered the most interesting problems in the microelectronics design in a
wide range of topics variety. The programme of the Conference included prepared invited
presentations, oral presentations and poster sessions. The best presentations have been
carefully chosen as the contents of this book. It is the reflection of issues and views
debated at the Conference and the summary of technical assessments and results presented.
The Conference was organised by the programme committee, and all the papers presented
at the Conference have been reviewed by at least two members of programme committee.
The best papers of each session were chosen for this book by the session chairmen and
then reviewed again by the following members of International Steering Committee:
Prof. A. Napieralski, Technical University ofL6di, Poland
Dr Z. Ciota, Technical University ofL6di, Poland
Prof. A. Martinez, Laboratoire d' Analyse et d' Architecture des Systemes du Centre
National de la Recherche Scientifique, France
Prof. G. De Mey, University of Gent, Belgium
Prof. J. Cabestany, Universitat Politecnica de Catalunya, Spain
Prof. A. De Vos, University of Gent, Belgium
Prof. R. Ubar, Tallinn Technical University, Estonia
Dr J. L. Noullet, Institut National des Sciences Appliquees de Toulouse, France
Prof. M. Glesner, Technische Hochschule, Darmstadt, Germany
Prof. W. Kuimicz, Warsaw University of Technology, Poland
Prof. A. Handkiewicz, Poznan University of Technology, Poland
Dr J. M. Moreno, Universitat Politecnica de Catalunya, Spain
This book groups together the selected papers (only 25% of presented paper have been
chosen) written by experts in technology, modelling, analogue and digital filters design,
signal processing, neural network design, thermal design, design methodologies,
electromagnetic compatibility, sensors and actuators, advanced testing methods and
reliability, power devices and Smart-Power modules. The special emphasis has been laid
on the educational aspects of all presented problems. The book is divided into the
following 5 parts:
1. Analog Circuit Design
(coordinated by Prof. G. De Mey and Dr Z. Ciota)
2. Power Devices & Thermal Aspects
(coordinated by Prof. A. Napieralski and Prof. G. De Mey)
3. Microsystems & Neural Networks
(coordinated by Prof. A. Martinez and Prof. J. Cabestany)
4. Design Methodologies
(coordinated by Prof. J. Cabestany and Dr J.L. Noullet)
5. Advanced Trends in Microelectronics Education
(coordinated by Dr J.L. Noullet and Dr Z. Ciota)
We hope that you will find this book interesting and it will be useful aid for all students
and engineers involved in VLSI circuits design.
 Preface xi
The Conference was sponsored by Poland Section IEEE - CAS Chapter, INCO-Copernicus
SYTIC Project, Polish State Committee for Scientific Research, ESPRIT-BARMINT
Program and Australasian Association for Engineering Education.
For all those who came together to share ideas, and for all the prospective readers, we
hope that this publication will excite engineers and students to search new ideas.

L6dz, Poland Andrzej NAPIERALSKI

L6di, Poland Zygmunt CIOTA
Toulouse, France Augustin MARTINEZ
Gent, Belgium Gilbert DE MEY
Barcelona, Spain Joan CABESTANY
 PART I

Analog Circuits
Design
 1
MULTICHANNEL LOW NOISE, LOW
POWER ANALOGUE READOUT CHIP
FOR SILICON STRIP DETECTORS
Wojciech Bialas, Wladyslaw D2lbrowski,
Pawel Grybos and Marek Idzik
Faculty ofPhysics and Nuclear Techniques
University of Mining and Metallurgy
al. Mickiewicza 30. 30-059 Krak6w
POLAND

ABSTRACT
A prototype 16-channel readout chip for a X-ray detection system using silicon strip
detectors is presented. The chip has been designed as a full custom ASIC for the AMS
1.2f./1T1 CMOS process. Single channel of the circuit consists of low noise. medium speed
preamplifier. followed by a shaper and a discriminator. Such a system allows to use
silicon strip detectors in the single photon counting mode. Key design issues and
optimization of critical design parameters are discussed. Measurement results from the
successfully manufactured prototype are presented and discussed.
1. INTRODUCTION
A silicon strip detector is an array of reverse biased junction diodes made on high
resistivity silicon substrate. The diodes are shaped as strips with pitch from 20llm to
200llm according to the geometrical requirements of an experiment. A particle crossing
the active detector region loses its energy and produces short current pulses on the strips in
the vicinity of particle trajectory. When silicon strip detectors are used for X-ray detection
in applications like X-ray imaging or X-ray diffractometry they work in the single photon
counting mode. An effective use of these sensors requires a special fast, low noise readout
electronics which essentially can be designed only as full custom VLSI ASICs. In this
paper we present the design and the measurement results of a prototype multichannel chip
optimized for readout of silicon strip detectors for a Roentgen diffractometer.
In the presented design a number of issues relevant for most analogue VLSI circuits, like
low-noise, low-power and matching, is addressed carefully. A current signal generated in
a strip of silicon detector by a single photon is amplified, shaped and applied to
a comparator providing I-bit yes/no information. For such a signal processing scheme
a sufficient separation between the noise level and the signal level is required in order to
provide a high detection efficiency and a limited noise count rate. For a system with a first
order band-pass filter of time constant T the noise count rate at the comparator output is
given by Rice formula in [1].
4 Analog Circuits Design

(1)

where V, is the comparator threshold, Vn is the RMS value of noise at the comparator input
andfnO is zero crossing frequency which equals to 0.29/r [2]. In order to keep the noise
count rate at a negligible level, i.e. 10-3 count/sec or below, it is required for a typical
detection system that the comparator threshold is set at least at a level of 3 Vn • On the other
hand the comparator threshold should be low enough to provide full efficiency for real
signal counts. Since the signal is smeared by noise then another margin of 3 Vn for the
threshold setting is needed. Then, taking into account some additional effects like
fluctuation of charge generated in silicon detectors, charge division between neighbouring
strips and matching of amplifier gain and comparator offset for a multichannel system, one
arrives to a conclusion that a signal-to-noise ratio of lOis required at minimum. In order
to make a multichannel system practical one requires that a common discrimination
threshold is applied for all channels. Therefore the matching of single channel parameters
in a multichannel chip is an absolutely critical issue for such a system.
A typical energy of X-rays used in Roentgen diffractometry is 8 keY. A photon of this
energy when absorbed in the depletion region of a silicon diode produces of about 2000
electron-hole pairs which are collected by the readout strips within the time period of
about 20 ns [3, 4]. In an ideal situation one expects the signal at the amplifier output to be
proportional to the charge collected in the detector. Thus for such a system we define the
charge gain as the ratio (output voltage signal)/(input charge) and the equivalent noise
charge (ENC) as the charge applied to the input in a form of a short S-Iike current pulse
which gives at the output a signal amplitude equal to the RMS value of noise Vn •

2. SINGLE CHANNEL CIRCUIT

The block diagram and full circuit implementation of a single channel is shown in
Figure 1. It comprises four blocks: preamplifier, shaping amplifier, differential stage and
discriminator.

(b)
Figure 1. The single channel ofthe chip: a) block diagram, b) full circuit diagram.
 Multichannel Low Noise, Low Power Analogue Readout Chip 5
for Silicon Strip Detectors

The preamplifier stage is the charge sensitive amplifier which integrates the current input
signal from a silicon strip detector into a voltage signal. The signal at the preamplifier
output is a step function with exponential decay defined be the long time constant of the
feedback loop. The voltage signal is applied to a band-pass filter which provides two
functions: (i) shaping of a short pulse according to timing requirements and (ii) limiting of
the bandwidth to a minimum corresponding to the signal spectral density in order to keep
minimum noise bandwidth.
In order to optimize the noise performance of the chip, the preamplifier and the shaper is
made in a single ended configuration. For such a system and a relatively short peaking
time, when the IIf noise is negligible, the equivalent noise charge ENC expressed in
number of electrons is given as [5]:

ENe =
(2)
where:
Ci" is the total input capacitance including the gate capacitance of the input transistor,
silicon detector capacitance and any stray capacitance introduced by connection between
the silicon strip detector and the readout chip,

V; is the spectral density of the equivalent input voltage noise, dominated by the
equivalent input noise of the input transistor,

i: is the spectral density of the equivalent input current noise dominated by the thermal
noise of the feedback resistor and the shot noise of the detector leakage current,
Tp is the peaking time, i.e. the time at which the signal at the filter output reaches the
maximum; for a simple CR-RC band-pass filter peaking time Tp is equal to the time
constant of the filter r= RC
Fv and Fi are constants dependent on the filter type.
Two immediate observations based on the formula (2) are important for optimization of
the front-end circuit, namely: (i) contribution of the input voltage noise to ENC is
proportional to the total input capacitance and inversely proportional to the square root of
the peaking time and (ii) contribution of the input current noise is independent of the input
capacitance and proportional to the square root of the peaking time. One can now optimize
the front-end system for various requirements and constrains given by a particular
application. If for example there is no constraint on the peaking time, like for any low rate
experiment, one can find an optimum peaking time for given other parameters, voltage and
current noise spectral densities and detector capacitance. However, in many applications
high counting rate capability is a serious requirement which has to taken into account as
a limitation for the peaking time. Another parameter driven strongly by applications is the
detector capacitance. Thus, in most cases in order to minimize ENC one has to focus on
the optimization of the input transistor.
Equivalent input voltage noise of a MOSFET is inversely proportional to the
transconductance gm' In strong inversion gm is proportional to ~(W I L)1d ' where
W. L. fa are the width, length and drain current of the input transistor respectively. To
minimize the voltage noise WIL and fa should be as large as possible, however, in order to
find an optimum WIL one has to take into account some other effects. First of all
a minimum L allowed by a given technology should not be used in order to avoid
6 Analog Circuits Design

an excess noise due to short channel effects (6). For a given gate length L the input
capacitance of the transistor will increase proportionally to the gate width W and will
contribute to the total input capacitance in the formula (2).
For our design we have chosen peaking time of 550ns based on requirements concerning
the maximum X-ray intensity per single strip. Taking into account the foreseen detector
capacitance of about 2 pF per strip and all other aspects discussed above we have chosen
a PMOS of WIL=1500f..lml1.5f..lm as an input device. For the chosen peaking time one can
expect some extra contribution to ENC due to IIf noise which is neglected in formula (2).
Since it is well known that IIf noise is usually lower for PMOS transistors compared to
NMOS transistors we have used a PMOS one although this solution has a small drawback
due to a higher current required for a given transconductance.
The preamplifier and the shaper are based on a folded cascode configuration which
provides good open loop gain and bandwidth as well as good power supply rejection ratio.
The last aspect is quite important since due to noise reason the preamplifier has to be
design in a single ended configuration. The currents in the preamplifier stage and in the
shaper are controlled by the external resistors RPRE and RSH (see Figure I) from several
f..lA to several hundred f..lA. This solution gives a possibility to control the gain, the peaking
time and the power consumption and to optimize in some range the settings for different
detector capacitances.
Another drawback of the single ended configurations of the preamplifier and the shaper
results in significant channel-to-channel and chip-to-chip offset variation Therefore AC
coupling is implemented between these two stages as well as between the shaper and the
comparator. Furthermore the rest of the circuit, after the shaper, is designed in a fully
differential mode. The differential pair after the shaper has a gain close to I and is
basically used for providing a differential threshold to the comparator. The differential
scheme for setting the comparator threshold allows to use the circuit either for positive or
for negative input signal polarity. Providing the gain in the preamplifier and shaper are
high enough the chosen solution allows us to apply a common threshold for all
16 channels in the chip and keep the comparator offset variation negligible compared to
noise.
3. LAYOUT
The circuit is laid out as 16-channel chip with 90 f..lm pitch. The length of the chip
including input and output bond pads is equal 2.5 mm. All bond pads for the bias and
control lines are placed on both sides of the chip. Along each channel a number of small
probe pads has been placed which provide a capability of probing the signal after each
stage of the circuit. In front of each preamplifier a test capacitor of 50 iF is placed. The test
signals are distributed from two calibration pads to every second channel from each. This
scheme provides a possibility for testing the cross-talk between channels.

4. TEST RESULTS
The chip has been manufactured successfully in the AMS 1.2f..lm process. The basic
functionality of a single channel is illustrated in the Figure 2. The measurements were
performed using the internal calibration circuitry. Applying a voltage step signal Vin to the
calibration capacitor C1 which is connected to the input of every channel is equivalent to
an injection of the charge Qin = C Vin . Figure 2 shows the signal wave forms measured at
the output of the shaper and at the output of the discriminator for an input signal of 1600
electrons. The upper plot shows results of measurements with a digital scope in the
averaging mode, while the lower plot shows the result of a single shot measurements
accumulated over some time period. The intensity of the shaper output wave form gives
 Multichannel Low Noise, Low Power Analogue Readout Chip 7
for Silicon Strip Detectors

a rough estimate of signal-to-noise ratio of about 20. Precise measurements of noise give
a typical ENC value of about 50 electrons RMS for zero detector capacitance. For the
nominal bias currents the power consumption is below 3 mW/channel.
-1._110 UI 2.01 oao UI '.110lIII0 UI

A.
- .-
-
\
.1-............- j
--- '--""" . ---- -....... ..... ............
-

-+-+-.~.~
,
'-

--
-

r----
.

_ ....

- ---'--

, 2.0 aoo 7.aaaaa UI

-1.Oaaoa UI UI

J
"- ....
--
..........-:

.+-+-t-.

Figure 2. Response ofthe shaper and comparator for an input charge of 1600 electrons
applied via the test capacitor. Comparator output measured for a threshold of300 m V.
Time scale: 1 ps/div. Vertical scale: 200 mV/div for the shaper output and 1V/div
for the comparator output.
The behavior of the shaper for different values of the bias current has been found as
expected from simulation. Measured shaper parameters (gain and peaking time) as
a function of the bias current are listed in Table 1. This feature offers a possibility to
minimize the noise for different detector capacitances by choosing a proper peaking time.
Table 1. Measured charge sensitivity and peaking time as a function ofthe shaper current.
Current in the shaper [J.lAI Charge sensitivity [mV/fCI Peaking time [nsl
40 650 800
60 580 550
80 400 400
8 Analog Circuits Design

5. CONCLUSIONS
A low noise, low power ASIC for readout of silicon strip detectors has been designed and
manufactured successfully. The experimental results obtained form the prototype are in
good agreement with the designed parameters. It has been shown that using the concept
employed in the presented circuit one can design a very low noise front-end circuitry for
silicon strip detectors allowing for efficient detection of X-ray photons of energy as low as
8 keY in the single photon counting mode.
REFERENCES
[I] S. O. Rice, Mathematical Analysis of Random Noise, Bell System Teeh. 1., 24 (1945), p.55
[2] H. Spieler, Power Requirements for Frontend Electronics in the Silicon Tracker, SCIPP
Report 91128, September, 1991.
[3] 1. Kemmer and G. Lutz, New detector concepts, Nuel. Instr. and Meth. Vol. A 253, (1987),
p.365.
[4] w. Dabrowski, P. Grybos and M. Idzik, Study of spatial resolution and efficiency of silicon
strip detectors with different readout schemes, Nuel. Instr. and Meth. Vol. A 356 (1995),
p.241.
[5] E. Gatti and P.F. Manfredi, Processing the signal from solid-state detectors in elementary
particle physics, La Revista del Nuovo Cimento, Vol. 9, No.1, 1986.
[6] W. Dabrowski et al. Noise measurements on radiation hardned CMOS tranSistor, SCIPP
Report 91194, University of California, Santa Cruz, 1991.
 2
DESIGN METHODOLOGY FOR CURRENT
CONVEYOR BASED CONTINUOUS-TIME
FIELD-PROGRAMMABLE ANALOG ARRAY
Richard Grisel, Christophe Premont,
Nacer Abouchi and Jean-Pierre Chante
CPE Lyon. LISA CNRS EP 0092
43 Bd. du II Nov. 1918. BP 2077 Villeurbqnne
FRANCE

ABSTRACT
A design methodology for continuous time Field-Programmable Analog Array (FPAA) is
presented. After introducing the key features of FPAA and dealing with design issues
related to continuous-time applications. we present the elementary cell of the proposed
FPAA. This cell is based on current conveyors and designed for high-frequency
applications. Two examples are presented: a high-frequency amplifier and a high-
frequency multiplier.
1. INTRODUCTION
One defines a Field-Programmable Analog Array (FPAA) as an integrated circuit which
allows, by the mean of flexible programmability facilities, to implement analog functions.
If we consider the recent market increase of programmable devices such as PALs, EPLDs,
FPGAs for the digital counterpart, it is obvious to say that, for real-time signal processing
applications, FPAA design should provide electronics designers with a very efficient and
powerful tool.
A FPAA consists of programmable analogue elementary cells which can be interconnected
by the mean of programmable interconnections. The reconfigurable cell has to perform
a set of functions, in order to provide the flexibility, with good electrical performance.
The set of functions is classically defined as: amplifier, comparator, multiplier, voltage-
controlled oscillator.
Different previous approaches are operational amplifier-based and operate at limited
bandwidth (100kHz) and have limited linearity due to the use of MOSFET based switches.
The proposed approach tries to cope with these two major hints, by the use of a current
conveyor based elementary cell for the reconfigurable analogue block, allowing
high-frequency continuous-time applications, and being programmable as a pass or
no-pass switch by tuning its bias sources.
10 Analog Circuits Design

2. DESIGN ISSUES FOR FIELD-PROGRAMMABLE ANALOG

ARRAY
Design of analog applications requires attention to several parameters like noise level,
distortion, dynamic range, etc. The two major key features for the design of an efficient
and useful analog array are the programming of the elementary functions and the
reconfigurable topology. The elementary cell function is set by changing some parameters
like the value of programmable transconductances and capacitors. The programmed cell
has to perform its function with good electrical performance offering wide parameter
range for a flexible analog array. Two different approaches, the continuous-time or the
switched approach, can be considered. These two techniques do not offer the same
trade-offs between performance and parameter range for the programming.
The switched approach consists of switched-capacitors or switched-current which are
digitally controlled and provide a wide parameter range but with frequencies limited to
few hundred kilohertz because ofthe Shannon theorem [1,2,3).
The continuous-time approach offers a lower parameter range compared to the switched
approach, but with particular design techniques, like the use of current conveyor for
example, some very efficient analog blocks can be developed for the analog array. In the
previous works [4,5] using the continuous-time approach for the design of
field-programmable analog array, the performance of the circuit were limited by the use of
both Op-Amp based design and analog switches, preventing high-frequency operation of
the circuits.
A new methodology for the elementary analog cell design is introduced, and a specific
approach for the interconnection of the cells without the use of switches in the signal path
is addressed. Several properties of current conveyors are used to achieve both a high-
frequency operation of the elementary cells and a local interconnection scheme. The
configuration of the circuit, the programming of the functions and of the topology require
the used of bit registers and control voltages. Some registers are converted with a digital to
analog converter to provide control voltages to set the values of the programmable
transconductances and capacitors of the array [6-7). The control voltages are locally
stored, near the element to program, on a capacitor which needs to be periodically
refreshed. New configuration techniques have been recently addressed, as for example the
use of EEPROM non-volatile analog memory. This technique has been thoroughly studied
in [8). The present paper focuses on the design of the analog cell, and the configuration of
the analog array and its architecture are not addressed.

3. ELEMENTARY ANALOG CELL

The analog elementary cell of the array, presented in Figure I, consists of a four MOSFET
transconductor and two conveyor-based I-V converters. The four MOSFET
transconductor [9-10] is a highly linear differential, programmable transconductor with
response given by:

(1)

Parameters 11 and Cox are respectively the average carrier mobility in the channel and the
gate oxide capacitance per unit area. L and W are respectively the length and the width of
the transistor. Relation (1) is true if the conditions V., V2 < min [VCl - Vp Vcl - Vr ] are
verified. (Vt is the threshold voltage for a MOS transistor)
 Design Methodology for Current Conveyor Based II
Continuous-Time Field-Programmable Analog Array

.....-...---""-_os.

TC'j
Figure 1. Elementary Analog Cell.
A current conveyor (CC) [II] is a three terminal device which operates in such a way that
if a voltage is applied to the input terminal Y, an equal potential will appear on the input
terminal X. In a similar fashion, an input current forced into terminal X will result in
an equal current flowing into terminal Z.
The two CCs of Figure I perform the I-V conversion. Virtual grounds at nodes X have to
be created in order to make the four MOSFET transconductor to behave linearly. The
currents IXp and IXn flowing into the X nodes produce the voltages VSp and VS n . Using
equation (I) the following relation is derived:

(2)

The difference (Vc I-Vc2) is used to perform a four-quadrant analog multiplier and to
control the polarity of the output signal. The load resistor and capacitor are used to
perform either an amplifier or an integrator or a low-pass first order filter.
The programmable resistor RI is a CMOS resistor with transistors operating in ohmic
region [12], in order to have both good linearity and parameter range. The programmable
capacitor is based on a capacitive multiplier thoroughly described in [13].
The design of an analog processing application is performed by cascading two CCs. The
output port Z of the first CC is connected to the input port Y of the second CC. If several
CCs port Z are connected to the same node, then the output currents are added and
converted in voltage with a resistor before going to the next stage.
As explained in the next section, a CC is biased by two current sources, the positive and
the negative one. A CC can perform interconnection between cells, simulating a pass
switch or a non-pass switch by turning on or off respectively its two bias current
sources [14]. The CC'S equivalent input impedance is modified with the polarisation.

4. mGH-FREQUENCY CONSIDERATIONS
In the presented analog elementary cell the two current conveyors are used to transfer the
current from the four MOSFET transconductor to the load. The frequency limitation is
due to the current transfer IzlIx. The schematic of a current conveyor, presented in
Figure 2, shows that the current transfer between terminal X and Z is performed by two
current mirrors M6-M8 and M7-M9. The voltage at node Y is copied to node X using the
12 Analog Circuits Design

current mirrors MI-M2 and M3-M4. The transistors MIO, Mil, MI2 and M13 are the
bias current sources controlled by two voltages Vp and Vn.

Ml0
v~

Yo---4I ....._oZ

Figure 2. Schematic ofthe current conveyor

It can be shown that for the elementary current mirror of Figure 3, the approximate first
order transfer functions is:
l-s Cgdl
1 IlNgml gml
gm; . ;.o.;lC;";~l,; dl; ; )
our =gVolff = --gm,--I+S(C+Cgdl + Cgsl+Cgdl + ..
g gmt gm,g

with g = _1_+ gds z and C = Cdbz + CL

RL

Figure 3.A simple current mirror.

gml, gm2 are the transconductance of transistor MI and M2 respectively. Cg d2 and Cdb2
are the gate-drain capacitance and drain-bulk capacitance of transistor M2 respectively.
Cgs I is the gate-source capacitance of transistor MI.
The main limitation are due to the Cgd2 capacitor and to the load impedance of the current
mirror. If low RI and CI value are used, referring to Figure I, and reasonable W/L ratios
are used to limit the Cgd2 value, a high-frequency operation can be achieved. The current
conveyor has been implemented using AMS O.8um CMOS technology previously used
in [15] and chosen for its stability as far as process parameters are concerned. HSpice
simulations have been carried out and a bandwidth greater then IOOMHz for the current
transfer is achieved using the W/L ratios given in Figure 2.
 Design Methodology for Current Conveyor Based 13
Continuous-Time Field-Programmable Analog Array

S. RESULTS
The frequency response of the elementary cell (see Figure 4) used as an amplifier with
a 20mV input signal for a voltage gain of 10 achieves IOMhz bandwidth (-3dB cut-off
frequency).
200m ~. - - - - - - - - - - - - - - - - - - - - - -...._ • . I . . . . . . . . . .. .

1?5", ~ ......• :.... .•.......•. . .•..••..•..•. :.~ .•...•.~

ISO", ;.. •••••.••:.... • ••.• •••• ••• • ••. '" .•••••• . : ...•••
\ ~
t
l75. !- ......• ~.... or.. ...... 1,... . ..... '1. • • • • • • • I.. . . ....... ::!
~ ~

.'.......:....
lGO. ~ ...... ~... .o • -t ........
. ; . . . . . . . .; . ••• .; .....•..i • ••••

?5", r·······~···· ······i······· ...... ~ ..... \.~

SO'IM r' ··.. ·.:~.l····, '.: .:1' '.·,'1'.'.:1' '·'.,,'1' ..1' ';~;I'I~( "-:":"111" ~'''I :.1' ···I·I·I:~
O.r-----:----:-----:----:----.........:.....:.....;......:.:
-20 .. .. .. , ... ..
' .. .'. ..
--'10 .. .. . .. .'.. . . . .. . .
. .............
-'0 .. '-i .. .
',' .. ',' .
-80
..,.. III .. , .., .. III ...... ...... . ',
III
II
10
II
100 "' Ik
1111111
10k
1111111
lOOk
1111111
I.
1111111
10.
hut! Clog) 100.
OUtput voltage: magnitude (upper curve), phase (lower curve)
Figure 4. Ccs based amplifier.
A frequency doubler can be performed using the four MOSFET transconductor as
an analog multiplier and two input signals with the same frequency. Figure 5 shows that
the circuit is able to perform well as a frequency doubler at 80MHz, for two 40MHz input
signals.
-
10.0,- ..........
(.\. .., . ..., . ':'j'"'-'
;~~ ... :
...... : ...... -: ;.. .... ,,~ .. ';' ..... -: .... ;""': ..... -: ......
.' \ .
20.0 :- ........ ;'- ...; \ ... i····· ./.. . \·i· ..........~... f..··F\. ....;........}_ .......\-:;
j . '. . i . \. . i . I.. • i . '..

_20':';,~~:: ;./:::: ::.. :."::<:: :.ft_·::·.. ·.·. :·;\~·:·. ;f(:·:·~:·~:: :\;~~:: :~~t:::.:.:.:.::~
-10.0 :...\~.: ....... .I . . . . ~_/ •• , I, . I. :'1._; 1_·, •• I.... ,_0", .' I. ,
I: =::tr::::./\.-:::. t\::::lI\:·.·.:/\·.·.·.·.Ji':~:
: .' ~ I: I I~: : ! ' ·'····I ..
:::{\.·.·.·.:-.·.t\~
t'
'-·i···\'····I··: \. ·· ... : .. ·).····:r· "",' . ·I ....
.':', I:',i·····I··
':!• • • • • • •,•••••
:: ..·;-
,,:
... I '1 I: I I: :1 , I '. I:, I: , :1
1 -
...
'·····r·· .,, ....•.............
• 1 '. ....
~ ...... I. ...•... 1..•.•...
\.
1
"..., •
j ••••
.. "
',.
t·· .•, . . . . '\-
•
2 :..; ..... ~ .. ~.... : .~ .. ;.: ..... '•. ~ ......... i ...... :I J.....:. ~ •.;.: •. .••• .l.. ..... ~
l·.l '. ,. \ .' ., 1
•

,
t

.. '. .' \j \'.. i

o ~ .... ·IO~'.1. .•. '2~'" .... i~~ .. "-t~~ .. 's~,,' . ,."v. ?~~ . : 8~~ .. ~~~ . '10:"
o l.",. (I,.. ) 100

Input signal (upper curve), output signal (lower curve)

Figure 5: Ccs basedfrequency doubler.

6. CONCLUSION
A new approach for designing analog elementary cell for field-programmable analog array
has been described. Its major improvement, compared to the previous works, is the use of
14 Analog Circuits Design

current conveyors in order to achieve a wide range of analog functions operating at high
frequency (80 MHz). The key performance feature of the proposed approach is the
current-mode processing which seems to provide attractive solution for wide bandwidth
capability.

ACKNOWLEDGEMENTS
The authors wish to thank the Scientific and Technology Cooperation Service of Ottawa
French embassy in Canada for supporting part of this work.

REFERENCES
[1] A Bratt and 1. Macbeth, Design and Implementation of a Field Programmable Analogue
Array, FPGA'96, pp. 88-93, 1996.
[2] H.W. Klein, The EPAC Architecture: An Expert Cell Approach to Field Programmable Analog
Array, FPGA'96, pp. 94-98,1996.
[3] S.T. Chang, B.R. Hayes-Gill and C.J. Paul, Multi-function Blockfor a Switched Current Field
Programmable Analog Array, MWSCAS'96, Ames (Iowa), 1996.
[4] K.F.E. Lee and P.G. Gulak, A CMOS Field-Programmable Analog Array, ISSCC Digest of
technical papers, pp. 186-188, Feb. 1991.
[5] K.F.E. Lee, Field-Programmable Analog Arrays Based on MOS Transconductors,
Ph.D. Thesis, University ofTonronto, 1995.
[6] K.W. Current and M.E. Hurlston, A Bi-Directionnal Current-Mode CMOS Multiple Valued
Logic Memory Circuit, Proc. of the Inter. Symposium on Multiple Valued Logic, pp. 196-202,
May 1991.
[7] B. Hochet, V. Peiris, S. Abdo and M.J. Declercq, Implementation of a Learning Kohonen
Neuron Based on a New Multilevel Storage Technique, IEEE Journal of Solid State Circuits,
Vol. 26, No.3, pp. 262-267, Mar. 1991.
[8] A Thomsen and M.A Brooke, Low Control Voltage Programming of Floating Gate MOSFETs
and Applications, IEEE Trans. on Circuits and Systems, Vol. 41, No.6, pp. 443-451,
June 1994.
[9] Z. Czamul, Novel MOS Resistive Circuit for syntheSiS of Fully Integrated Continuous-time
Filters, IEEE Trans. on Circuits and Systems, Vol. 33, No.7, pp. 718-721, July 1986.
[l0] S.T. Dupuie and M. Ismail, High Frequency CMOS transconductors, in Analog IC Design:
The Current-Mode approach, Peter Peregrinus Ltd., 1990.
[ll] AS. Sedra, C.W. Roberts and F. Gohh, The Current-Conveyors: History, progress and new
results, lEE Proc., Vol. 137, Pt. G, No.2, April 1990.
[12] J. Silva-Martinez, M. Steyaert and W. Sansen, High-Performance CMOS Continuous-time
filters, Kluwer Academic Publishers, 1993.
[13] C. Premont, R. Grisel, N. Abouchi and J. P. Chante, A Current Conveyor based Capacitive
Multiplier, MIXDES'97, Poznan, Poland, pp. 81-84, June 1997.
[l4] C. Premont, N. Abouchi, R. Grisel and J.P. Chante, A Current Conveyor based High-Frequency
Analog Switch, IEEE Trans. on Circuits and Systems, to be published in Dec. 1997.
[15] F. Bergouignan, N. Abouchi, R. Grisel, G. Caille and J. Caranana, Designs of a Logarithmic
and Exponential Amplifier Using Current Conveyor, ICECS'96, Rodos, Greece, pp. 61-62,
Oct. 1996.
 3
RF AMP IC FOR OPTICAL DISC PLAYER
ICbun-Sup Kim, IGea-Ok Cbo,
IYong-Hwan Kim and lHang-Sup Song
JASIC Center, Corporate Technical Operations
Samsung Electronics Co., Ltd
KOREA
2 University
ofIllinois,
Urbana,IL
USA

ABSTRACT
This paper describes RF amp integrated circuit for the read channel of optical disc
players that include CD and DVD demanding the equalizer and the decision block. A RF
equalizer is composed of boosting jilter to slim the data pulse and Ijh order Bessel jilter to
attenuate high frequency noise. Boosting jilter emphasize 6dB at 6MHz and maintains the
linear phase within 1% by pole zero cancellation method. An important advantage is
tuneability of the boosting and frequency characteristics by controlling the
transconductance, therefore, it is enough to cover the tolerance offabrication. The high
frequency noise due to equalization is attenuated by Ijh order Bessel jilter which has the
slope of over -42dBloct. Measured jitter keeps within 4.5n. The decision block includes
one bit analog to digital converter and the low pass jilters to detect asymmetry and
asymmetric amp. As the analog and the digital signals are mixed on a chip, the main
ground and the NMOS bulk ground, DGND and AGND respectively, are separated. Also,
it is laid out very carefully to minimize the digital noise in analog block. Implemented in
a O.Bum CMOS n-well technology, the RF equalizer and the decision block chip occupy
3000um x lOOOum. Power consumption from a single 5V is around 300mW

1. INTRODUCTION
Recently the optical disc has become a choice of high-density storage medium for
CD-ROM or digital video applications. In addition to its high capacity, the high-speed
read capability has also become a key requirement for the success of the optical medium.
The reading speed of the compressed MPEG-encoded data stored in optical discs, for
example, is vital for their use. In the new DVD (Digital Versatile Disc) format, the stomge
capacity is 4.7GB and the data spectrum covers the frequency range of 5 to 6MHz.
Compared to the existing CD format of 720kHz, the DVD needs higher speed electronics
to compensate for high-frequency loss.
This paper presents the RF amp chip for high-speed optical discs and DVD players. The
RF equalizer is integrated on one chip with decision block. It is implements by a Bessel
filter with two symmetric zeros using a tuneable gm-C topology. The 9th order filter is
16 Analog Circuits Design

chosen to effectively suppress high-frequency noise and to mllllmlze the jitter.

The decision block is made up of the one bit analog to digital converter and asymmetric
circuits for compensating the slice level.

2. RF EQUALIZER ARCIDTECTURE
The block diagram of the RF equalizer is shown in Figure I. The signal, picked up by the
laser diode assembly, is amplified by a current controlled amplifier (CCA) that sets the
signal level to 1Vpp. The equalizer, which uses a gm-C topology, boosts the
high-frequency gain using two zeros. This high-frequency boosting is often called as pulse
slimming because the pulse looks slimmer after boosting. As is true in the most data
communications system, the linear phase characteristic of the filter is of prime interest to
maintain the pulse timing integrity. The transfer function of two symmetric zeros,
H(S) = (I+As)(l-Bs), is used so that a linear phase characteristic can be obtained.

-8>-
The high-frequency out-of-band noise is attenuated with a -42dB/oct slope.

--00 Boost filter 9th order lDr HFE<~O

Micom
--~~----~----------~

Figure 1. Block diagram ofthe RF equalizer.

The 9th order Bessel lowpass filter is made of a cascade of 5 biquad stages [I, 2].
All biquad stages are scaled so that all node voltages can have the same magnitude at the
cut-off frequency of the equalizer. Figure 2 illustrates the first biquad stage with 2 zeros.
It can be implemented fully differentially, but the single-ended version is shown. The
tuneable gm is drawn as a resistor in the figure. The cut-off frequency of the equalizer is
controlled by gm) while the boost amount is set independently by gm2 of the frequency
control. The tuneable range is set to cover the process variation of about ±10%.
gml

CI

gml

VI

V2

C3 +

Figure. 2. First biquad with 2 zeros.

The simulation results of the gain and group delay of the DVD equalizer are shown
in Figure 3.
 RF Amp Ie for Optical Disc Player 17

...:oo.'I'I!Q•

....... .; ...... . ...... .~ ..... , .;........~ ....... ;........:..... .

•• : .: : .:
•.•.... ; •..•••. ! •.•••.••••.•.. : ...••••. ;..... ,

....... ...... .
~ ....... ....... !........ .......
~ ~ ~

.. . ..

.$' , ..............;............... ....... ; ....... ~ ........ ~........ !....... ':' ... .

"01."

"lIlt
..-
·\oii_.·,;,;··..··iiI··-jii·.··""··~·ioiiio!iioio:piii··..··ioi··ip·:·iii··,;,;··ioi··..:··ioii··,;,;··~··:~··iii··oiii··ijo·:·;,;;··.··,;;,··op':'·iii··;f'··
,....
II1II1%( :l.loI
L'....
.>Mot

(a) Gain

''''''01
.SL.,r-.....,-....,..--,.--""'---'--""""-"""'''''''''''''''''''''''--'
.so. ·······;······.1.·······i···· .. ·j· .. ·····~·······!·······t···· .. ·:··· . ··t····
..... ......!....... ~........ ~ ....... j........l ....... \...... , ....... :....... '- .... .
')0)0

••
110. ....... :....... ;........:....... :........f··· .. ·· \........~ ....... :.. .
'00. ....... ~ ....... ;... ·.... f.... ·.. :..... ' 'r ....... i· .......f· ...... !............ .
".SoIo
'.U",",
'1lOIC

(b) Group Delay

Figure 3. Simulated gain and delay ofthe DVD equalizer.

The boost and group delay deviation over 1 to 6MHz for DVD are designed to be 6dB at
6MHz and within 2ns, respectively.

3. THE DECISION BLOCK

The decision block diagram is shown in Figure 4. The decision block converts the
equalized analog signal to digital data. It affects the perfonnance of the digital signal
process as following system.
18 Analog Circuits Design

the equalized signal

Ibit analog
RFE~ RFEQI to digital
sedll Digital data
To DSP

vf converter I
I
I
I Low pass filter
AS,(
- Asymmetric amp
to detect
I asymmetry

Figure 4. Block diagram a/the decision

3.1. One bit analog to digital converter

The block diagram of one bit analog to digital converter is shown in Figure 5. The input is
compared with the feedback voltage, vf, that is compensated on the asymmetric error of
digital data by the low pass filter and asymmetric amp. The comparator shown in Figure 6
has a hysteresis of about 70 mV to make it insensitive to noise [3]. Also, the main ground
and the nmos bulk ground, DGND and AGND, are separated, respectively. It is laid out
very carefully to minimize digital noise in analog block.
VDD
RFEQI output buffer
----1+
Comparator
vf

AGND
DGND

Figure 5. One bit analog to digital converter

M3 M4

HFEQI ----i t--_M_l_ _---"'n""m:..;,.o~s..::.b""UI"'k--r_ _ _M2_---1~ vf

---j I---'-AGND
DGND
Figure 6. Comparator with a hysteresis.

3.2. Auto asymmetry control amplifier

It is shown in Figure 7 that sets the adequate signal level for the comparator. The reference
voltage VC is adjustable to appropriate level to adapt to versatile disc mode.
 RF Amp IC for Optical Disc Player 19

Vout = VC + 80k(VC-asy)/20k (1)

vc vf
,>--.,.-- To comparator
ASY
20k

80k
Figure 7. Auto asymmetry control amplifier Circuits

4. EXPERIMENTAL RESULTS
Fabricated using a O.8um CMOS technology that uses the n-well p-type substrate, the RF
amp chip occupies 3000um x 1000um. Measured gain and phase responses of the
equalizer shown in Figure 8 are very close to the simulated ones shown in Figure 3.
AEF 0.0 dll'" ATT 1.0 at!
S dB;' - n .. 1"""

.... w
2000

v .. w
300
k ... z

kM z
--- ...............
~
~

START 0.1 .... HZ STOP J.2.0 MHZ

(a) Gain
REF 0.0 <;115m
"0· ,/ ·s."

-
"-
'" "-
'" "" "- ~

'"
.... w
1.00 KHZ
"-.. '-...

"'\. '\.
vew
300 kHz:

"- '\.
SWP 30 m. STOP 12.0 MH%

(b) Phase
Figure 8. Measured gain and phase of the equalizer.

When the gain at 6MHz is boosted up to 6dB, the group delay deviation is within 2 to 4ns
and the phase linearity is measured to be less than 1%. The jitter measured within 4.5ns is
very good for optical disc player. Figure 9 shows the frequency tuning characteristics of
the RF equalizer.
Other documents randomly have
different content
a book. Then the volume was replaced and the tedious search went
on.
Suddenly some slight noise—Gherkins was sure he did not make it; it
may have been Bunter in the dining-room—seemed to catch the ear
of the kneeling man.
"Wot's that?" he gasped, and his startled face swung round into view.
"Hands up!" said Lord Peter, and switched the light on.
The second man made one leap for the dining-room door, where a
smash and an oath proclaimed that he had encountered Bunter. The
kneeling man shot his hands up like a marionette.
"Gherkins," said Lord Peter, "do you think you can go across to that
gentleman by the bookcase and relieve him of the article which is so
inelegantly distending the right-hand pocket of his coat? Wait a
minute. Don't on any account get between him and my pistol, and
mind you take the thing out very carefully. There's no hurry. That's
splendid. Just point it at the floor while you bring it across, would
you? Thanks. Bunter has managed for himself, I see. Now run into
my bedroom, and in the bottom of my wardrobe you will find a bundle
of stout cord. Oh! I beg your pardon; yes, put your hands down by all
means. It must be very tiring exercise."
The arms of the intruders being secured behind their backs with a
neatness which Gherkins felt to be worthy of the best traditions of
Sexton Blake, Lord Peter motioned his captives to sit down and
despatched Bunter for whisky-and-soda.
"Before we send for the police," said Lord Peter, "you would do me a
great personal favour by telling me what you were looking for, and
who sent you. Ah! thanks, Bunter. As our guests are not at liberty to
use their hands, perhaps you would be kind enough to assist them to
a drink. Now then, say when."
"Well, you're a gentleman, guv'nor," said the First Burglar, wiping his
mouth politely on his shoulder, the back of his hand not being
available. "If we'd a known wot a job this wos goin' ter be, blow me if
we'd a touched it. The bloke said, ses 'e, 'It's takin' candy from a
baby,' 'e ses. 'The gentleman's a reg'lar softie,' 'e ses, 'one o' these
'ere sersiety toffs wiv a maggot fer old books,' that's wot 'e ses, 'an' ef
yer can find this 'ere old book fer me,' 'e ses, 'there's a pony fer yer.'
Well! Sech a job! 'E didn't mention as 'ow there'd be five 'undred
fousand bleedin' ole books all as alike as a regiment o' bleedin'
dragoons. Nor as 'ow yer kept a nice little machine-gun like that 'andy
by the bedside, nor yet as 'ow yer was so bleedin' good at tyin' knots
in a bit o' string. No—'e didn't think ter mention them things."
"Deuced unsporting of him," said his lordship. "Do you happen to
know the gentleman's name?"
"No—that was another o' them things wot 'e didn't mention. 'E's a
stout, fair party, wiv 'orn rims to 'is goggles and a bald 'ead. One o'
these 'ere philanthropists, I reckon. A friend o' mine, wot got inter
trouble onct, got work froo 'im, and the gentleman comes round and
ses to 'im, 'e ses, 'Could yer find me a couple o' lads ter do a little
job?' 'e ses, an' my friend, finkin' no 'arm, you see, guv'nor, but wot it
might be a bit of a joke like, 'e gets 'old of my pal an' me, an' we
meets the gentleman in a pub dahn Whitechapel way. W'ich we was
ter meet 'im there again Friday night, us 'avin' allowed that time fer ter
git 'old of the book."
"The book being, if I may hazard a guess, the Cosmographia
Universalis?"
"Sumfink like that, guv'nor. I got its jaw-breakin' name wrote down on
a bit o' paper, wot my pal 'ad in 'is 'and. Wot did yer do wiv that 'ere
bit o' paper, Bill?"
"Well, look here," said Lord Peter, "I'm afraid I must send for the
police, but I think it likely, if you give us your assistance to get hold of
your gentleman, whose name I strongly suspect to be Wilberforce
Pope, that you will get off pretty easily. Telephone the police, Bunter,
and then go and put something on that eye of yours. Gherkins, we'll
give these gentlemen another drink, and then I think perhaps you'd
better hop back to bed; the fun's over. No? Well, put a good thick coat
on, there's a good fellow, because what your mother will say to me if
you catch a cold I don't like to think."
So the police had come and taken the burglars away, and now
Detective-Inspector Parker, of Scotland Yard, a great personal friend
of Lord Peter's, sat toying with a cup of coffee and listening to the
story.
"But what's the matter with the jolly old book, anyhow, to make it so
popular?" he demanded.
"I don't know," replied Wimsey; "but after Mr. Pope's little visit the
other day I got kind of intrigued about it and had a look through it. I've
got a hunch it may turn out rather valuable, after all. Unsuspected
beauties and all that sort of thing. If only Mr. Pope had been a trifle
more accurate in his facts, he might have got away with something to
which I feel pretty sure he isn't entitled. Anyway, when I'd seen—what
I saw, I wrote off to Dr. Conyers of Yelsall Manor, the late owner——"
"Conyers, the cancer man?"
"Yes. He's done some pretty important research in his time, I fancy.
Getting on now, though; about seventy-eight, I fancy. I hope he's
more honest than his nephew, with one foot in the grave like that.
Anyway, I wrote (with Gherkins's permission, naturally) to say we had
the book and had been specially interested by something we found
there, and would he be so obliging as to tell us something of its
history. I also——"
"But what did you find in it?"
"I don't think we'll tell him yet, Gherkins, shall we? I like to keep
policemen guessing. As I was saying, when you so rudely interrupted
me, I also asked him whether he knew anything about his good
nephew's offer to buy it back. His answer has just arrived. He says he
knows of nothing specially interesting about the book. It has been in
the library untold years, and the tearing out of the maps must have
been done a long time ago by some family vandal. He can't think why
his nephew should be so keen on it, as he certainly never pored over
it as a boy. In fact, the old man declares the engaging Wilberforce
has never even set foot in Yelsall Manor to his knowledge. So much
for the fire-breathing monsters and the pleasant Sunday afternoons."
"Naughty Wilberforce!"
"M'm. Yes. So, after last night's little dust-up, I wired the old boy we
were tooling down to Yelsall to have a heart-to-heart talk with him
about his picture-book and his nephew."
"Are you taking the book down with you?" asked Parker. "I can give
you a police escort for it if you like."
"That's not a bad idea," said Wimsey. "We don't know where the
insinuating Mr. Pope may be hanging out, and I wouldn't put it past
him to make another attempt."
"Better be on the safe side," said Parker. "I can't come myself, but I'll
send down a couple of men with you."
"Good egg," said Lord Peter. "Call up your myrmidons. We'll get a car
round at once. You're coming, Gherkins, I suppose? God knows what
your mother would say. Don't ever be an uncle, Charles; it's frightfully
difficult to be fair to all parties."

Yelsall Manor was one of those large, decaying country mansions

which speak eloquently of times more spacious than our own. The
original late Tudor construction had been masked by the addition of a
wide frontage in the Italian manner, with a kind of classical portico
surmounted by a pediment and approached by a semi-circular flight
of steps. The grounds had originally been laid out in that formal
manner in which grove nods to grove and each half duly reflects the
other. A late owner, however, had burst out into the more eccentric
sort of landscape gardening which is associated with the name of
Capability Brown. A Chinese pagoda, somewhat resembling Sir
William Chambers's erection in Kew Gardens, but smaller, rose out of
a grove of laurustinus towards the eastern extremity of the house,
while at the rear appeared a large artificial lake, dotted with numerous
islands, on which odd little temples, grottos, tea-houses, and bridges
peeped out from among clumps of shrubs, once ornamental, but now
sadly overgrown. A boat-house, with wide eaves like the designs on a
willow-pattern plate, stood at one corner, its landing-stage fallen into
decay and wreathed with melancholy weeds.
"My disreputable old ancestor, Cuthbert Conyers, settled down here
when he retired from the sea in 1732," said Dr. Conyers, smiling
faintly. "His elder brother died childless, so the black sheep returned
to the fold with the determination to become respectable and found a
family. I fear he did not succeed altogether. There were very queer
tales as to where his money came from. He is said to have been a
pirate, and to have sailed with the notorious Captain Blackbeard. In
the village, to this day, he is remembered and spoken of as Cut-throat
Conyers. It used to make the old man very angry, and there is an
unpleasant story of his slicing the ears off a groom who had been
heard to call him 'Old Cut-throat.' He was not an uncultivated person,
though. It was he who did the landscape-gardening round at the
back, and he built the pagoda for his telescope. He was reputed to
study the Black Art, and there were certainly a number of astrological
works in the library with his name on the fly-leaf, but probably the
telescope was only a remembrance of his seafaring days.
"Anyhow, towards the end of his life he became more and more odd
and morose. He quarrelled with his family, and turned his younger
son out of doors with his wife and children. An unpleasant old fellow.
"On his deathbed he was attended by the parson—a good, earnest,
God-fearing sort of man, who must have put up with a deal of insult in
carrying out what he firmly believed to be the sacred duty of
reconciling the old man to this shamefully treated son. Eventually,
'Old Cut-throat' relented so far as to make a will, leaving to the
younger son 'My treasure which I have buried in Munster.' The parson
represented to him that it was useless to bequeath a treasure unless
he also bequeathed the information where to find it, but the horrid old
pirate only chuckled spitefully, and said that, as he had been at the
pains to collect the treasure, his son might well be at the pains of
looking for it. Further than that he would not go, and so he died, and I
dare say went to a very bad place.
"Since then the family has died out, and I am the sole representative
of the Conyers, and heir to the treasure, whatever and wherever it is,
for it was never discovered. I do not suppose it was very honestly
come by, but, since it would be useless now to try and find the
original owners, I imagine I have a better right to it than anybody
living.
"You may think it very unseemly, Lord Peter, that an old, lonely man
like myself should be greedy for a hoard of pirate's gold. But my
whole life has been devoted to studying the disease of cancer, and I
believe myself to be very close to a solution of one part at least of the
terrible problem. Research costs money, and my limited means are
very nearly exhausted. The property is mortgaged up to the hilt, and I
do most urgently desire to complete my experiments before I die, and
to leave a sufficient sum to found a clinic where the work can be
carried on.
"During the last year I have made very great efforts to solve the
mystery of 'Old Cut-throat's' treasure. I have been able to leave much
of my experimental work in the most capable hands of my assistant,
Dr. Forbes, while I pursued my researches with the very slender clue
I had to go upon. It was the more expensive and difficult that Cuthbert
had left no indication in his will whether Münster in Germany or
Munster in Ireland was the hiding-place of the treasure. My journeys
and my search in both places cost money and brought me no further
on my quest. I returned, disheartened, in August, and found myself
obliged to sell my library, in order to defray my expenses and obtain a
little money with which to struggle on with my sadly delayed
experiments."
"Ah!" said Lord Peter. "I begin to see light."
The old physician looked at him enquiringly. They had finished tea,
and were seated around the great fireplace in the study. Lord Peter's
interested questions about the beautiful, dilapidated old house and
estate had led the conversation naturally to Dr. Conyers's family,
shelving for the time the problem of the Cosmographia, which lay on
a table beside them.
"Everything you say fits into the puzzle," went on Wimsey, "and I think
there's not the smallest doubt what Mr. Wilberforce Pope was after,
though how he knew that you had the Cosmographia here I couldn't
say."
"When I disposed of the library, I sent him a catalogue," said Dr.
Conyers. "As a relative, I thought he ought to have the right to buy
anything he fancied. I can't think why he didn't secure the book then,
instead of behaving in this most shocking fashion."
Lord Peter hooted with laughter.
"Why, because he never tumbled to it till afterwards," he said. "And
oh, dear, how wild he must have been! I forgive him everything.
Although," he added, "I don't want to raise your hopes too high, sir,
for, even when we've solved old Cuthbert's riddle, I don't know that
we're very much nearer to the treasure."
"To the treasure?"
"Well, now, sir. I want you first to look at this page, where there's a
name scrawled in the margin. Our ancestors had an untidy way of
signing their possessions higgledy-piggledy in margins instead of in a
decent, Christian way in the fly-leaf. This is a handwriting of
somewhere about Charles I's reign: 'Jac: Coniers.' I take it that goes
to prove that the book was in the possession of your family at any
rate as early as the first half of the seventeenth century, and has
remained there ever since. Right. Now we turn to page 1099, where
we find a description of the discoveries of Christopher Columbus. It's
headed, you see, by a kind of map, with some of Mr. Pope's monsters
swimming about in it, and apparently representing the Canaries, or,
as they used to be called, the Fortunate Isles. It doesn't look much
more accurate than old maps usually are, but I take it the big island
on the right is meant for Lanzarote, and the two nearest to it may be
Teneriffe and Gran Canaria."
"But what's that writing in the middle?"
"That's just the point. The writing is later than 'Jac: Coniers's'
signature; I should put it about 1700—but, of course, it may have
been written a good deal later still. I mean, a man who was elderly in
1730 would still use the style of writing he adopted as a young man,
especially if, like your ancestor the pirate, he had spent the early part
of his life in outdoor pursuits and hadn't done much writing."
"Do you mean to say, Uncle Peter," broke in the viscount excitedly,
"that that's 'Old Cut-throat's' writing?"
"I'd be ready to lay a sporting bet it is. Look here, sir, you've been
scouring round Münster in Germany and Munster in Ireland—but how
about good old Sebastian Munster here in the library at home?"
"God bless my soul! Is it possible?"
"It's pretty nearly certain, sir. Here's what he says, written, you see,
round the head of that sort of sea-dragon:

"Hic in capite draconis ardet perpetuo Sol.

Here the sun shines perpetually upon the Dragon's Head.
 THE DRAGON'S HEAD
Liber V.
1099
DE NOVIS INSVLIS,
quomodo, quando, & per quem
illæ inuentæ sint.
Christophorus Columbus natione Genuensis, cùm
diu in aula regis Hispanorum deuersarus fuisset,
animum induxit, ut hactenus inacceslias orbis
partes peragraret. Pet à rege, utuoto suo non
deesset, futurum sibi & toti Hisp

"Rather doggy Latin—sea-dog Latin, you might say, in fact."

"I'm afraid," said Dr. Conyers, "I must be very stupid, but I can't see
where that leads us."
"No; 'Old Cut-throat' was rather clever. No doubt he thought that, if
anybody read it, they'd think it was just an allusion to where it says,
further down, that 'the islands were called Fortunatæ because of the
wonderful temperature of the air and the clemency of the skies.' But
the cunning old astrologer up in his pagoda had a meaning of his
own. Here's a little book published in 1678—Middleton's Practical
Astrology—just the sort of popular handbook an amateur like 'Old
Cut-throat' would use. Here you are: 'If in your figure you find Jupiter
or Venus or Dragon's head, you may be confident there is Treasure in
the place supposed.... If you find Sol to be the significator of the
hidden Treasure, you may conclude there is Gold, or some jewels.'
You know, sir, I think we may conclude it."
"Dear me!" said Dr. Conyers. "I believe, indeed, you must be right.
And I am ashamed to think that if anybody had suggested to me that
it could ever be profitable to me to learn the terms of astrology, I
should have replied in my vanity that my time was too valuable to
waste on such foolishness. I am deeply indebted to you."
"Yes," said Gherkins, "but where is the treasure, uncle?"
"That's just it," said Lord Peter. "The map is very vague; there is no
latitude or longitude given; and the directions, such as they are, seem
not even to refer to any spot on the islands, but to some place in the
middle of the sea. Besides, it is nearly two hundred years since the
treasure was hidden, and it may already have been found by
somebody or other."
Dr. Conyers stood up.
"I am an old man," he said, "but I still have some strength. If I can by
any means get together the money for an expedition, I will not rest till
I have made every possible effort to find the treasure and to endow
my clinic."
"Then, sir, I hope you'll let me give a hand to the good work," said
Lord Peter.
Dr. Conyers had invited his guests to stay the night, and, after the
excited viscount had been packed off to bed, Wimsey and the old
man sat late, consulting maps and diligently reading Munster's
chapter "De Novis Insulis," in the hope of discovering some further
clue. At length, however, they separated, and Lord Peter went
upstairs, the book under his arm. He was restless, however, and,
instead of going to bed, sat for a long time at his window, which
looked out upon the lake. The moon, a few days past the full, was
riding high among small, windy clouds, and picked out the sharp
eaves of the Chinese tea-houses and the straggling tops of the
unpruned shrubs. 'Old Cut-throat' and his landscape-gardening!
Wimsey could have fancied that the old pirate was sitting now beside
his telescope in the preposterous pagoda, chuckling over his riddling
testament and counting the craters of the moon. "If Luna, there is
silver." The water of the lake was silver enough; there was a great
smooth path across it, broken by the sinister wedge of the boat-
house, the black shadows of the islands, and, almost in the middle of
the lake, a decayed fountain, a writhing Celestial dragon-shape,
spiny-backed and ridiculous.
Wimsey rubbed his eyes. There was something strangely familiar
about the lake; from moment to moment it assumed the queer
unreality of a place which one recognises without having ever known
it. It was like one's first sight of the Leaning Tower of Pisa—too like its
picture to be quite believable. Surely, thought Wimsey, he knew that
elongated island on the right, shaped rather like a winged monster,
with its two little clumps of buildings. And the island to the left of it,
like the British Isles, but warped out of shape. And the third island,
between the others, and nearer. The three formed a triangle, with the
Chinese fountain in the centre, the moon shining steadily upon its
dragon head. "Hic in capite draconis ardet perpetuo——"
Lord Peter sprang up with a loud exclamation, and flung open the
door into the dressing-room. A small figure wrapped in an eiderdown
hurriedly uncoiled itself from the window-seat.
"I'm sorry, Uncle Peter," said Gherkins. "I was so dreadfully wide
awake, it wasn't any good staying in bed."
"Come here," said Lord Peter, "and tell me if I'm mad or dreaming.
Look out of the window and compare it with the map—Old Cut-
throat's 'New Islands.' He made 'em, Gherkins; he put 'em here.
Aren't they laid out just like the Canaries? Those three islands in a
triangle, and the fourth down here in the corner? And the boat-house
where the big ship is in the picture? And the dragon fountain where
the dragon's head is? Well, my son, that's where your hidden
treasure's gone to. Get your things on, Gherkins, and damn the time
when all good little boys should be in bed! We're going for a row on
the lake, if there's a tub in that boat-house that'll float."
"Oh, Uncle Peter! This is a real adventure!"
"All right," said Wimsey. "Fifteen men on the dead man's chest, and
all that! Yo-ho-ho, and a bottle of Johnny Walker! Pirate expedition
fitted out in dead of night to seek hidden treasure and explore the
Fortunate Isles! Come on, crew!"

Lord Peter hitched the leaky dinghy to the dragon's knobbly tail and
climbed out carefully, for the base of the fountain was green and
weedy.
"I'm afraid it's your job to sit there and bail, Gherkins," he said. "All the
best captains bag the really interesting jobs for themselves. We'd
better start with the head. If the old blighter said head, he probably
meant it." He passed an arm affectionately round the creature's neck
for support, while he methodically pressed and pulled the various
knobs and bumps of its anatomy. "It seems beastly solid, but I'm sure
there's a spring somewhere. You won't forget to bail, will you? I'd
simply hate to turn round and find the boat gone. Pirate chief
marooned on island and all that. Well, it isn't its back hair, anyhow.
We'll try its eyes. I say, Gherkins, I'm sure I felt something move, only
it's frightfully stiff. We might have thought to bring some oil. Never
mind; it's dogged as does it. It's coming. It's coming. Booh! Pah!"
A fierce effort thrust the rusted knob inwards, releasing a huge spout
of water into his face from the dragon's gaping throat. The fountain,
dry for many years, soared rejoicingly heavenwards, drenching the
treasure-hunters, and making rainbows in the moonlight.
"I suppose this is 'Old Cut-throat's' idea of humour," grumbled
Wimsey, retreating cautiously round the dragon's neck. "And now I
can't turn it off again. Well, dash it all, let's try the other eye."
He pressed for a few moments in vain. Then, with a grinding clang,
the bronze wings of the monster clapped down to its sides, revealing
a deep square hole, and the fountain ceased to play.
"Gherkins!" said Lord Peter, "we've done it. (But don't neglect bailing
on that account!) There's a box here. And it's beastly heavy. No; all
right, I can manage. Gimme the boat-hook. Now I do hope the old
sinner really did have a treasure. What a bore if it's only one of his
little jokes. Never mind—hold the boat steady. There. Always
remember, Gherkins, that you can make quite an effective crane with
a boat-hook and a stout pair of braces. Got it? That's right. Now for
home and beauty.... Hullo! what's all that?"
As he paddled the boat round, it was evident that something was
happening down by the boat-house. Lights were moving about, and a
sound of voices came across the lake.
"They think we're burglars, Gherkins. Always misunderstood. Give
way, my hearties—

"A-roving, a-roving, since roving's been my ru-i-in,

I'll go no more a-roving with you, fair maid."

"Is that you, my lord?" said a man's voice as they drew in to the boat-
house.
"Why, it's our faithful sleuths!" cried his lordship. "What's the
excitement?"
"We found this fellow sneaking round the boat-house," said the man
from Scotland Yard. "He says he's the old gentleman's nephew. Do
you know him, my lord?"
"I rather fancy I do," said Wimsey. "Mr. Pope, I think. Good evening.
Were you looking for anything? Not a treasure, by any chance?
Because we've just found one. Oh! don't say that. Maxima reverentia,
you know. Lord St. George is of tender years. And, by the way, thank
you so much for sending your delightful friends to call on me last
night. Oh, yes, Thompson, I'll charge him all right. You there, doctor?
Splendid. Now, if anybody's got a spanner or anything handy, we'll
have a look at Great-grandpapa Cuthbert. And if he turns out to be
old iron, Mr. Pope, you'll have had an uncommonly good joke for your
money."
An iron bar was produced from the boat-house and thrust under the
hasp of the chest. It creaked and burst. Dr. Conyers knelt down
tremulously and threw open the lid.
There was a little pause.
"The drinks are on you, Mr. Pope," said Lord Peter. "I think, doctor, it
ought to be a jolly good hospital when it's finished."
 THE PISCATORIAL FARCE OF THE
STOLEN STOMACH
"What in the world," said Lord Peter Wimsey, "is that?"
Thomas Macpherson disengaged the tall jar from its final swathings
of paper and straw and set it tenderly upright beside the coffee-pot.
"That," he said, "is Great-Uncle Joseph's legacy."
"And who is Great-Uncle Joseph?"
"He was my mother's uncle. Name of Ferguson. Eccentric old boy. I
was rather a favourite of his."
"It looks like it. Was that all he left you?"
"Imph'm. He said a good digestion was the most precious thing a
man could have."
"Well, he was right there. Is this his? Was it a good one?"
"Good enough. He lived to be ninety-five, and never had a day's
illness."
Wimsey looked at the jar with increased respect.
"What did he die of?"
"Chucked himself out of a sixth-story window. He had a stroke, and
the doctors told him—or he guessed for himself—that it was the
beginning of the end. He left a letter. Said he had never been ill in his
life and wasn't going to begin now. They brought it in temporary
insanity, of course, but I think he was thoroughly sensible."
"I should say so. What was he when he was functioning?"
"He used to be in business—something to do with ship-building, I
believe, but he retired long ago. He was what the papers call a
recluse. Lived all by himself in a little top flat in Glasgow, and saw
nobody. Used to go off by himself for days at a time, nobody knew
where or why. I used to look him up about once a year and take him
a bottle of whisky."
"Had he any money?"
"Nobody knew. He ought to have had—he was a rich man when he
retired. But, when we came to look into it, it turned out he only had a
balance of about five hundred pounds in the Glasgow Bank.
Apparently he drew out almost everything he had about twenty years
ago. There were one or two big bank failures round about that time,
and they thought he must have got the wind up. But what he did with
it, goodness only knows."
"Kept it in an old stocking, I expect."
"I should think Cousin Robert devoutly hopes so."
"Cousin Robert?"
"He's the residuary legatee. Distant connection of mine, and the only
remaining Ferguson. He was awfully wild when he found he'd only
got five hundred. He's rather a bright lad, is Robert, and a few
thousands would have come in handy."
"I see. Well, how about a bit of brekker? You might stick Great-Uncle
Joseph out of the way somewhere. I don't care about the looks of
him."
"I thought you were rather partial to anatomical specimens."
"So I am, but not on the breakfast-table. 'A place for everything and
everything in its place,' as my grandmother used to say. Besides, it
would give Maggie a shock if she saw it."
Macpherson laughed, and transferred the jar to a cupboard.
"Maggie's shock-proof. I brought a few odd bones and things with
me, by way of a holiday task. I'm getting near my final, you know.
She'll just think this is another of them. Ring the bell, old man, would
you? We'll see what the trout's like."
The door opened to admit the housekeeper, with a dish of grilled
trout and a plate of fried scones.
"These look good, Maggie," said Wimsey, drawing his chair up and
sniffing appreciatively.
"Aye, sir, they're gude, but they're awfu' wee fish."
"Don't grumble at them," said Macpherson. "They're the sole result of
a day's purgatory up on Loch Whyneon. What with the sun fit to
roast you and an east wind, I'm pretty well flayed alive. I very nearly
didn't shave at all this morning." He passed a reminiscent hand over
his red and excoriated face. "Ugh! It's a stiff pull up that hill, and the
boat was going wallop, wallop all the time, like being in the Bay of
Biscay."
"Damnable, I should think. But there's a change coming. The glass is
going back. We'll be having some rain before we're many days
older."
"Time, too," said Macpherson. "The burns are nearly dry, and there's
not much water in the Fleet." He glanced out of the window to where
the little river ran tinkling and skinkling over the stones at the bottom
of the garden. "If only we get a few days' rain now, there'll be some
grand fishing."
"It would come just as I've got to go, naturally," remarked Wimsey.
"Yes; can't you stay a bit longer? I want to have a try for some sea-
trout."
"Sorry, old man, can't be done. I must be in Town on Wednesday.
Never mind. I've had a fine time in the fresh air and got in some
good rounds of golf."
"You must come up another time. I'm here for a month—getting my
strength up for the exams and all that. If you can't get away before I
go, we'll put it off till August and have a shot at the grouse. The
cottage is always at your service, you know, Wimsey."
"Many thanks. I may get my business over quicker than I think, and,
if I do, I'll turn up here again. When did you say your great-uncle
died?"
Macpherson stared at him.
"Some time in April, as far as I can remember. Why?"
"Oh, nothing—I just wondered. You were a favourite of his, didn't you
say?"
"In a sense. I think the old boy liked my remembering him from time
to time. Old people are pleased by little attentions, you know."
"M'm. Well, it's a queer world. What did you say his name was?"
"Ferguson—Joseph Alexander Ferguson, to be exact. You seem
extraordinarily interested in Great-Uncle Joseph."
"I thought, while I was about it, I might look up a man I know in the
ship-building line, and see if he knows anything about where the
money went to."
"If you can do that, Cousin Robert will give you a medal. But, if you
really want to exercise your detective powers on the problem, you'd
better have a hunt through the flat in Glasgow."
"Yes—what is the address, by the way?"
Macpherson told him the address.
"I'll make a note of it, and, if anything occurs to me, I'll communicate
with Cousin Robert. Where does he hang out?"
"Oh, he's in London, in a solicitor's office. Crosbie & Plump,
somewhere in Bloomsbury. Robert was studying for the Scottish Bar,
you know, but he made rather a mess of things, so they pushed him
off among the Sassenachs. His father died a couple of years ago—
he was a Writer to the Signet in Edinburgh—and I fancy Robert has
rather gone to the bow-wows since then. Got among a cheerful
crowd down there, don't you know, and wasted his substance
somewhat."
"Terrible! Scotsmen shouldn't be allowed to leave home. What are
you going to do with Great-Uncle?"
"Oh, I don't know. Keep him for a bit, I think. I liked the old fellow,
and I don't want to throw him away. He'll look rather well in my
consulting-room, don't you think, when I'm qualified and set up my
brass plate. I'll say he was presented by a grateful patient on whom I
performed a marvellous operation."
"That's a good idea. Stomach-grafting. Miracle of surgery never
before attempted. He'll bring sufferers to your door in flocks."
"Good old Great-Uncle—he may be worth a fortune to me after all."
"So he may. I don't suppose you've got such a thing as a photograph
of him, have you?"
"A photograph?" Macpherson stared again. "Great-Uncle seems to
be becoming a passion with you. I don't suppose the old man had a
photograph taken these thirty years. There was one done then—
when he retired from business. I expect Robert's got that."
"Och aye," said Wimsey, in the language of the country.

Wimsey left Scotland that evening, and drove down through the night
towards London, thinking hard as he went. He handled the wheel
mechanically, swerving now and again to avoid the green eyes of
rabbits as they bolted from the roadside to squat fascinated in the
glare of his head-lamps. He was accustomed to say that his brain
worked better when his immediate attention was occupied by the
incidents of the road.
Monday morning found him in town with his business finished and
his thinking done. A consultation with his ship-building friend had put
him in possession of some facts about Great-Uncle Joseph's money,
together with a copy of Great-Uncle Joseph's photograph, supplied
by the London representative of the Glasgow firm to which he had
belonged. It appeared that old Ferguson had been a man of mark in
his day. The portrait showed a fine, dour old face, long-lipped and
high in the cheek-bones—one of those faces which alter little in a
lifetime. Wimsey looked at the photograph with satisfaction as he
slipped it into his pocket and made a bee-line for Somerset House.
Here he wandered timidly about the wills department, till a uniformed
official took pity on him and enquired what he wanted.
"Oh, thank you," said Wimsey effusively, "thank you so much.
Always feel nervous in these places. All these big desks and things,
don't you know, so awe-inspiring and business-like. Yes, I just
wanted to have a squint at a will. I'm told you can see anybody's will
for a shilling. Is that really so?"
"Yes, sir, certainly. Anybody's will in particular, sir?"
"Oh, yes, of course—how silly of me. Yes. Curious, isn't it, that when
you're dead any stranger can come and snoop round your private
affairs—see how much you cut up for and who your lady friends
were, and all that. Yes. Not at all nice. Horrid lack of privacy, what?"
The attendant laughed.
"I expect it's all one when you're dead, sir."
"That's awfully true. Yes, naturally, you're dead by then and it doesn't
matter. May be a bit trying for your relations, of course, to learn what
a bad boy you've been. Great fun annoyin' one's relations. Always do
it myself. Now, what were we sayin'? Ah! yes—the will. (I'm always
so absent-minded.) Whose will, you said? Well, it's an old Scots
gentleman called Joseph Alexander Ferguson that died at Glasgow
—you know Glasgow, where the accent's so strong that even
Scotsmen faint when they hear it—in April, this last April as ever
was. If it's not troubling you too much, may I have a bob's-worth of
Joseph Alexander Ferguson?"
The attendant assured him that he might, adding the caution that he
must memorise the contents of the will and not on any account take
notes. Thus warned, Wimsey was conducted into a retired corner,
where in a short time the will was placed before him.
It was a commendably brief document, written in holograph, and was
dated the previous January. After the usual preamble and the
bequest of a few small sums and articles of personal ornament to
friends, it proceeded somewhat as follows:
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com