![Computer-Based Measurement Systems 9
quan = 1/Nx , (1.1)
where Nx is the result of a measurement of a quantity A.
The limit value of the absolute uncertainty A of measurement of a quantity A with
a digital instrument, e.g. a digital voltmeter, is specified by the manufacturer of the
instrument as a sum of two terms:
A = read Ax + range Arange , (1.2)
where Ad is the reading, Arange denotes the measurement range, read is the
uncertainty term representing a percentage of the reading not smaller than the
quantization error quan, and range is the uncertainty term representing a percentage
of the measurement range.
Example
In voltage measurements a voltmeter is used with the relative uncertainty:
read = 0.02% and range = 0.005% for the ranges of 10V and 100V. What is the
absolute uncertanity V of voltage measurement when we measure the voltage of
3.0V (Vx = 3.0V)?
When we used 10V-range of the voltmeter (Vrange = 10V) the absolute uncertainty is
V = [(Vread)2
+ (Vrange)2
]0.5
= [(read Vx)2
+ (range Vrange)2
]0.5
=
= [(0.02%3.0V)2
+ (0.005%10V)2
]0.5
= [(6 mV)2
+ (5 mV)2
]0.5
= 7.8 mV.
When we used 100V-range of the voltmeter (Vrange = 100V):
V = [(Vread)2
+ (Vrange)2
]0.5
= [(read Vx)2
+ (range Vrange)2
]0.5
=
= [(0.02%3.0V)2
+ (0.005%100V)2
]0.5
= [(6 mV)2
+ (50 mV)2
]0.5
= 51 mV.
Conclusion: In order to provide a small uncertainty the measuring range of an
instrument should be chosen as low as possible.
Manufacturers of instruments specify the uncertainty terms read and range in a
table presenting the values of these terms for different measurement ranges and
observation times. Usually the values of read and range are specified for the
observation times of 24 hours, 90 days, and one year.
When planning the measurement tasks to be executed by a measurement system we
can have an influence on many factors related to the measurement control and the
processing of measurement results, depicted schematically in Figure 1.5.](https://image.slidesharecdn.com/18506967-250601200143-b130b4a1/85/Measurement-Systems-And-Sensors-Second-Edition-2nd-Ed-Nawrocki-21-320.jpg)
![Computer-Based Measurement Systems 13
Measurement systems often use series transmission. The message transmission time
Ttr is a function of the rate R of data transfer (data rate) by the interface bus and the
size of the message, which is the size of the data frame or control frame, specified in
bits:
[bps]
[bits]
R
n
Ttr . (1.4)
Figure 1.7 Message transmission in a measurement system and the system dynamics.
The time Tprop of propagation of an electromagnetic (EMG) wave is a function of
the velocity v of propagation of the wave in the cable (or in air) and the length l of
the transmission line:
Tprop = l/v. (1.5)
If the length of the bus does not exceed 1 km, the time Tprop of propagation of
an electromagnetic wave carrying a measurement signal is usually shorter than the
transmission time Ttr. The role of the propagation time is sometimes underestimated
in the calculation of the time balance of a system. We usually note that the velocity
of propagation of electromagnetic waves, including light, in vacuum is 3 108
m/s,
the upper limit of signal propagation speed in nature. With this value of propagation
speed, an EMG wave will cover a distance of 1000 meters in a time of 3.3 µs. This,
however, only applies to propagation in vacuum. In a transmission line in the form
of an electrical cable or optical fiber the velocity of propagation of an
electromagnetic wave depends on the quality of the cable or otical fiber, and is
around 2 108
m/s. Transmission lines 1 km long are not unusual in measurement
systems. In an electrical cable or optical fiber of this length the propagation time of
an EMG wave will be about 5 µs.
Expressed in samples per second (sps), the sampling frequency S (i.e., number
of samples transmitted to a computer per unit of time), depends on the acquisition
time Taq and can`t be greater than 1/Taq:
S 1/Taq . (1.6)](https://image.slidesharecdn.com/18506967-250601200143-b130b4a1/85/Measurement-Systems-And-Sensors-Second-Edition-2nd-Ed-Nawrocki-25-320.jpg)
![Measurement Systems and Sensors
28
512-MB RAM. The compute module has a slot for an SD memory card with an
operating system. The SD memory card is also used for data storage. Raspberry
PI has also two USB connectors, an HDMI connector (usually used for
connecting a monitor), and an RJ45 connector for Ethernet.
2.3 COMPUTER ARCHITECTURE
The modular structure of modern PCs makes it possible to connect additional
circuits (e.g., boards, memory circuits), as well as peripheral devices to various
points of computer buses. The architecture of PCs has changed over time from the
multibus architecture [Figure 2.2(a)] used from 1992 to 2007 to the current
architecture [Figure 2.2(b)] used in computers manufactured since 2007. The
central component in the current architecture is North Chipset (Chipset 1),
controlled from the processor and controlling many buses and peripheral devices.
The point of joining the interface board to the computer is very important
with regard to the data processing rate. The further from the processor (in the
computer architecture) the interface board is, the slower the communication with
this board and the operation of the board itself will be. In Figure 2.2 a simplified
block diagram of the PC motherboard and the distribution of buses in the
computer are shown along with the arrangement of junctions provided by
constructors for joining additional or external devices. In Figure 2.2(a) and 2.2(b),
neither buses nor nonessential ports for the measurement system are shown, nor
are the E-IDE bus (for hard disks and CD), the SCSI bus, PS2 junctions for a
keyboard or a mouse, or junctions for a floppy disk. Additional computer boards
are joined to sockets inside the computer in the space provided for this purpose, in
the so-called slots. External devices are joined to junctions (sockets or connectors,
according to the type of interface) mounted into the computer casing.
The multibus architecture [Figure 2.2(a)] was used in computers
manufactured until around 2007. Computers with this architecture are still used in
many industrial and research measurement systems. The architecture shown in
Figure 2.2(a) has three buses: a Front Side Bus (FSB) for data exchange between
the processor and both the random access memory (RAM) and the cache memory,
a high data rate Peripheral Component Interconnect (PCI) bus, and a slow data
rate Industry Standard Architecture (ISA) bus.](https://image.slidesharecdn.com/18506967-250601200143-b130b4a1/85/Measurement-Systems-And-Sensors-Second-Edition-2nd-Ed-Nawrocki-40-320.jpg)
![Computer-Based Measurement Systems 33
• AT-CAN (by NI), the controller board of the measurement system with the
CAN bus with 1 or 2 ports;
• The I/O boards contain an analog-to-digital converter and a digital-to-
analog converter [e.g., NI 5102 (by NI)].
The chipset 2 circuit controls the USB, with a transmission rate up to 480
Mbps (USB 2.0). In these computer measurement systems in which a laptop is the
system controller, parameters of the Personal Computer Memory Card
International Association (PCMCIA) bus are essential.
The PCMCIA bus is set up in computers of this class and a similar PCI bus in
PCs. The PCMCIA ports were introduced in 1989 in order to connect additional
memory boards to the portable computer. Instead of the last Version 2.0 of the
PCMCIA standard, the Card Bus standard was introduced in 1994, with
parameters similar to the PCI bus. Essential differences between the PCMCIA
Version 2.0 and the Card Bus consist of the extension of data wordlengths from
16 bits (PCMCIA) to 32 bits, and in increasing clock frequency of the bus up to
33 MHz. In the new standard (Card Bus), the dimension of the PCMCIA boards
and 68-pin junctions to the bus are retained. The PCMCIA name is still applied
for both the bus and junctions and for computer boards fulfilling the conditions of
the Card Bus standard. The power supply voltage of PCMCIA boards is 3.3V.
The PCMCIA junction serves to join additional memory cards or additional hard
disks to the laptop. Moreover, for such junctions, one can connect the following
measuring devices in the form of the PCMCIA boards: the interface board (IEEE-
488, RS-232C, CAN), the DAQ (measuring board) with ADCs and DACs, the
modem board of the PSTN telephony, and the modem of the GSM cellular
telephony. The PCMCIA boards have the following standard dimensions: a length
of 85.6 mm, a width of 54 mm, and a thickness d different for three card types:
the PCMCIA type I—the thickness d = 3.3 mm; the PCMCIA type II—d = 5 mm;
and the PCMCIA type III—d = 10.5 mm.
At present, in modern laptops slots for PCMCIA boards are seldom installed.
2.4 UNIVERSAL SERIAL BUS
Measurement systems are usually very simple systems composed of one digital
instrument and one computer (see Chapter 7). For construction of such a system,
one can use the RS-232C serial interface installed in almost every PC. An
essential limitation of the system with this interface is a transmission rate not
higher than 20 kbps for transmission lines with a length of 15m, and not higher
than 115 kbps for systems with a short transmission line of 1.5m. Data
transmission in the computer system and in the computer-based measurement
system can be realized with a rate considerably higher than in the RS-232C
interface.](https://image.slidesharecdn.com/18506967-250601200143-b130b4a1/85/Measurement-Systems-And-Sensors-Second-Edition-2nd-Ed-Nawrocki-45-320.jpg)
![and requested to see the body of the Rosemere victim, which she immediately
identified as that of Maurice Greywood.
Strangely enough, the police throw doubts on this identification, although they
acknowledge that they have no other clue to go on. However, Mrs. Greywood, the
young man’s mother, has been sent for, and is expected to arrive to-morrow from
Maine, where she is spending the summer.
The people at the Rosemere are still foolishly trying to make a mystery of the
murder, and refuse all information [etc., etc.].
To Dr. Charles K. Fortescue from Dr. Frederic Cowper, Beverley, L. I.
Sunday Evening, August 13th.
Dear Charley:
No sooner had I read in to-day’s paper that the body found in the Rosemere had
been identified as that of Maurice Greywood, than I knew at once why you have
taken such an interest in poor May. I see now that you have suspected from the
first that the murdered man was not unknown to her, and your last letter,
describing her “friend,” proves to me beyond doubt that you were ignorant of
nothing but his name, for Greywood and no other answers exactly to that
description. How you found out what you did, I can’t imagine; but remembering
that your office window commands a view of the entrance to the building, I think
it possible that you may have seen something from that point of vantage, which
enabled you to put two and two together. But I wonder that I can feel any surprise
at your having discovered the truth, when the truth itself is unbelievable!! May
Derwent is incapable of killing any one—no matter what provocation she may have
had. She is incapable of a dishonourable action, and above all things incapable of
an intrigue. She is purity itself. I swear it. And yet what are the facts that confront
us? A man, known to have been her professed suitor, is found dead in a room
adjoining her apartment, dead with a wound through his heart—a wound, too,
caused by a knitting-needle or hat-pin, as you yourself testified! And before trying
to find out who killed him we must first think of some reasonable excuse for his
having been at the Rosemere at all. How strange that he should happen to go to
the building at the very time when May (who was supposed to be on her way to
Bar Harbor, mind you!) was there also. Who was he calling on, if not on her?
Luckily, no one as yet seems to have thought of her in connection with
Greywood’s death. My sister has, in fact, been wondering all day whom he could
have been visiting when he met his tragic fate. But, sooner or later, the truth will
become known, and then—? Even in imagination I can’t face that possibility.](https://image.slidesharecdn.com/18506967-250601200143-b130b4a1/85/Measurement-Systems-And-Sensors-Second-Edition-2nd-Ed-Nawrocki-86-320.jpg)
Measurement Systems And Sensors Second Edition 2nd Ed Nawrocki
- 1. Measurement Systems And Sensors Second Edition 2nd Ed Nawrocki download https://ebookbell.com/product/measurement-systems-and-sensors- second-edition-2nd-ed-nawrocki-37013934 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Measurement Systems And Sensors Second Edition Waldemar Nawrocki https://ebookbell.com/product/measurement-systems-and-sensors-second- edition-waldemar-nawrocki-61082530 Measurement Systems And Sensors Waldemar Nawrocki https://ebookbell.com/product/measurement-systems-and-sensors- waldemar-nawrocki-1130722 Sensors And Measurement Systems River Publishers Series In Electronic Materials And Devices 2nd Edition Lang Phd https://ebookbell.com/product/sensors-and-measurement-systems-river- publishers-series-in-electronic-materials-and-devices-2nd-edition- lang-phd-24210728 Advanced Interfacing Techniques For Sensors Measurement Circuits And Systems For Intelligent Sensors Boby George https://ebookbell.com/product/advanced-interfacing-techniques-for- sensors-measurement-circuits-and-systems-for-intelligent-sensors-boby- george-5868876
- 3. Phasor Measurement Units And Wide Area Monitoring Systems From The Sensors To The System 1st Edition Antonello Monti https://ebookbell.com/product/phasor-measurement-units-and-wide-area- monitoring-systems-from-the-sensors-to-the-system-1st-edition- antonello-monti-5601842 Measurement And Sensor Systems A Comprehensive Guide To Principles Practical Issues And Applications Alexander W Koch https://ebookbell.com/product/measurement-and-sensor-systems-a- comprehensive-guide-to-principles-practical-issues-and-applications- alexander-w-koch-48843222 Polarization Measurement And Control In Optical Fiber Communication And Sensor Systems Xiaojun James Chen https://ebookbell.com/product/polarization-measurement-and-control-in- optical-fiber-communication-and-sensor-systems-xiaojun-james- chen-47242520 Radiation Detection Systems Sensor Materials Systems Technology And Characterization Measurements Devices Circuits And Systems 2nd Edition https://ebookbell.com/product/radiation-detection-systems-sensor- materials-systems-technology-and-characterization-measurements- devices-circuits-and-systems-2nd-edition-34882140 Measurement Systems Application And Design 5th Edition Ernest O Doebelin https://ebookbell.com/product/measurement-systems-application-and- design-5th-edition-ernest-o-doebelin-5319378
- 5. Measurement Systems and Sensors Second Edition Waldemar Nawrocki naw_FM.indd iii naw_FM.indd iii 10/23/2015 3:16:27 PM 10/23/2015 3:16:27 PM
- 6. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the U.S. Library of Congress. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Cover design by John Gomes ISBN 13: 978-1-60807-932-2 © 2016 ARTECH HOUSE 685 Canton Street Norwood, MA 02062 All rights reserved. Printed and bound in the U nited States of America. N o part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the publisher. All terms mentioned in this book that are known to be trademarks or service marks have been appropriately capitalized. Artech House cannot attest to the accuracy of this information. Use of a term in this book should not be r egarded as affecting the validity of any trademark or service mark. 10 9 8 7 6 5 4 3 2 1 naw_FM.indd iv naw_FM.indd iv 10/23/2015 3:16:28 PM 10/23/2015 3:16:28 PM
- 7. V Contents Chapter 1 Computer-Based Measurement Systems 1 1.1 Configuration and Structure of Measurement Systems 1 1.2 Interface System 5 1.2.1 Interface System Meaning 5 1.2.2 Interface Bus 6 1.2.3 Interface Functions 7 1.3 Measurement Accuracy and Measurement System Dynamics 8 1.3.1 Accuracy of Measurement Systems 8 1.3.2 Measurement System Dynamics 10 1.4 Interface Protection 15 1.4.1 Interference in Measurement Instruments 15 1.4.2 Interference Induced in Transmission Line 20 Selected Bibliography 24 Chapter 2 Computers for Measurement Systems 25 2.1 Functions of Computer in Measurement Systems 25 2.2 Types of Computers for Measurement Systems 26 2.3 Computer Architecture 28 2.4 Universal Serial Bus 33 2.5 IEEE-1394 Serial Bus 40 Selected Bibliography 43 Chapter 3 Temperature Sensors 45 3.1. International Temperature Scale (ITS-90) 45 3.2 Resistance Sensors 47 3.2.1 Platinum Sensors 47 3.2.2 Thermistors 50 3.3 Thermocouples 53 3.4 Semiconductor Temperature Sensors 58 References 64 Chapter 4 Stress, Pressure and Acceleration Sensors 65 4.1 Mechanical Stresses and Pressure 65 4.2 Resistance Strain Gauges 67 4.3 Capacitive Gauges 71 4.4 Piezoelectric Sensors 72 4.5 Semiconductor Pressure Sensors 73 4.6 Accelerometers and Gyroscopes 77 4.6.1 Accelerometers 78
- 8. Measurement Systems and Sensors VI 4.6.2 Gyroscopes 81 References 85 Selected Bibliography 85 Chapter 5 Signal Conditioners 87 5.1 Voltage and Current Amplifiers 88 5.2 Voltage Conditioners 93 5.3 Conditioners for Temperature Sensors 95 5.4 Conditioners for Strain Gauges and Piezoelectric Sensors 98 5.5 Conditioners for Linear Position Sensors 99 References 101 Chapter 6 Digital-to-Analog and Analog-to-Digital Converters 103 6.1 Sampling and Quantizing 103 6.1.1 Sampling 103 6.1.2 Quantizing 106 6.2 Digital-to-Analog Converters 108 6.2.1 Parameters of Digital-to-Analog Converter 108 6.2.2 DACs with Resistor Dividers 110 6.2.3 DACs with PDM 115 6.2.4 Integrated DACs 117 6.2.5 Digital Controlled Potentiometers and Capacitors 119 6.3 Analog-to-Digital Converters 121 6.3.1 Analog-to-Digital Conversion Methods 121 6.3.2 Dual Slope Converters 122 6.3.3 Converters with Voltage-to-Frequency Conversion 127 6.3.4 A/D Converters with Successive Approximation Register 130 6.3.5 Flash Converters 132 6.3.6 Delta-Sigma A/D Converters 133 References 135 Selected Bibliography 135 Chapter 7 Measurement Systems with Serial Interface 137 7.1 Measurement Serial Interfaces an Overview 137 7.2 RS-232 Serial Interface System 138 7.2.1 General Description 138 7.2.2 Transmission in the RS-232C Interface System 140 7.2.3 RS-232C Interface Bus 145 7.2.4 Current Loop in the RS-232C Interface System 148 7.2.5 Null Modem Measurement System with the RS-232C Interface 149
- 9. VII Contents 7.3 Programming of Measurement System with the RS-232C Interface 153 7.3.1 Programming of the Null Modem System 153 7.3.2 ScopView Program for the Metex Multimeter 154 7.3.3 Thermo Program for Temperature Measurements 158 7.4 Measurement System with the RS-232C Interface and Modem 163 7.4.1 Modem 163 7.4.2 System with the RS-232C Interface and Telephone Modem 167 7.4.3 Programs for Data Transmission Control in a Distributed Measurement System 170 7.5 Other Serial Interface Systems 174 7.5.1 RS-449 and RS-530 Serial Interface Systems 174 7.5.2 RS-449 and RS-530 Standards for Serial Interface Circuits 177 7.5.3 A comparison of RS serial interface standards 180 7.6 Smart Sensors Interfaces 183 7.6.1 Smart Sensors 183 7.6.2 PROFIBUS Interface System 184 7.6.3 MicroLAN Interface System 187 7.7 Power Line Communication for Measurements 190 7.7.1 General Description of PLC 190 7.7.2 Communication Protocols for PLC 193 7.7.3 Data Acquisition System for Electricity Meters 194 References 195 Chapter 8 Wireless Measurement Systems 197 8.1 Wireless Transmission of Measurement Data 197 8.2 Radiomodem-Based Measurements Systems 198 8.2.1 Radio Channels and Radiomodems 198 8.2.2 Radiomodems in Measurement Systems 201 8.2.3 Measurement Systems with Radio Transmission: GSM-Based Versus Radiomodem-Based 202 8.2 Bluetooth Radio Link 204 8.3.1 ISM Frequency Bands for Short Range Communication 204 8.3.2 Bluetooth Standard – Basics 205 8.3.3 Applications of Bluetooth for Measures 209 8.4 IEEE-802.15.4 (ZigBee) Radio Link 213 8.4.1 General Information 213 8.4.2 IEEE 802.15.4 Transmission Protocols and Frames 214
- 10. Measurement Systems and Sensors VIII 8.4.3 ZigBee Network Topology and Devices 216 8.5 Other Wireless Transmission Systems 220 8.5.1 IrDA Infrared Link 220 8.5.2 WiFi Technology for Measurement Systems 224 8.5.3 Short Distance Wireless Transmission Systems – Comparation 226 References 228 Chapter 9 Measurement Systems with GSM and LTE 231 9.1 Wireless Transmission of Measurement Data 231 9.2 Measurement Systems with GSM-Based Data Transmission 232 9.2.1 GSM Mobile Phone Network 232 9.2.2 GSM-Based Data Transmission 235 9.3 GSM-Based Distributed Measurement Systems 242 9.4 UMTS, UDPA and LTE Telecommunication Systems 248 9.4.1 Universal Mobile Telecommunication System 248 9.4.2 HSPA and LTE 252 9.5 Mobile Stations 255 9.6 Positioning Systems 257 9.6.1 GPS Satellite Positioning System 257 9.6.2 GLONASS Positioning System 261 9.6.3 Galileo System and the Regional Positioning Systems: BeiDou, IRNSS and QZSS 263 9.6.4 Positioning System with UMTS 265 9.6.5 Applications of Positioning Systems 267 References 268 Selected Bibliography 268 Chapter 10 Measurement Systems with IEEE-488 Interface 269 10.1 IEEE-488 (IEC-625) Parallel Interface Standard 269 10.1.1 Parallel Interfaces 269 10.1.2 IEEE-488 Basic Specifications and Applications 270 10.1.3 Controller of IEEE-488 System 272 10.1.4 IEEE-488 Interface Bus and Cable 278 10.1.5 IEEE-488 Interface Functions 282 10.2 IEEE-488 Interface Messages and their Transfer 284 10.2.1 Interface Message Types 284 10.2.2 Remote Messages 286 10.2.3 Local Messages 290 10.2.4 Message Transfer in Handshake Mode 291 10.3 Enhancements in Measurement Systems with IEEE-488 interface 292 10.3.1 Enhancing Transfer Rates in Measurement Systems: HS488 Protocol 292
- 11. IX Contents 10.3.2 Increasing the Number of Devices in Measurement Systems 295 10.3.3 Distributed Measurement Systems with IEEE-488 Interface 297 10.4 IEEE-488 interface Function State Diagrams 301 10.4.1 Execution of Interface Functions 301 10.4.2 State Diagrams of Interface Functions 302 References 317 Chapter 11 Crate and Modular Measurement Systems 319 11.1 CAMAC System 319 11.1.1 Crate and Modular Measurement Systems an Overview 319 11.1.2 Organization of the CAMAC System 320 11.1.3 CAMAC Dataway 321 11.2 VXI Measurement System 322 11.2.1 General Specification 322 11.2.2 VXI Buses 324 11.3 PXI modular Measurement System 327 11.3.1 General Specification 327 11.3.2 PXI Bus 328 11.3.3 PXI System Modules 329 11.3.4 PXI-Express for Measurement Systems 331 11.3.5 PXI Measurement System Configuration 333 11.4 IEEE-1284 Interfaces in Measurement Systems 333 11.4.1 Bus of IEEE-1284 and Data Transmission 333 11.4.2 Measuring Systems with IEEE-1284 335 References 337 Chapter 12 LAN-Based Measurement Systems 339 12.1 Introduction 339 12.2 Ethernet Hardware 340 12.3 Ethernet Transfer Protocol 344 12.4 Ethernet-Based Measurement Systems 348 12.4.1 Ethernet-Based Measurement Systems with LAN/IEEE-488 Converters 348 12.4.2 Measurement Systems with LAN Interface 350 12.5 Internet-Based Measurement Systems and Systems with Embedded Web Servers 353 References 357 Selected Bibliography 358 Chapter 13 DAQ Boards and Virtual Instruments 359 13.1 Computer DAQ Boards 359
- 12. Measurement Systems and Sensors X 13.1.1 Structure and Functions of a DAQ board 359 13.1.2 DAQ Board Specifications 361 13.1.3 Selected DAQ Board Model Specifications 363 13.2 Virtual Instruments 367 13.3 Programming of Measurement Systems and Virtual Instruments 379 13.3.1 Software Development in the LabVIEW Environment 370 13.3.2 Programming of LAN-based and Internet-Based Measurement Systems 372 13.3.3 Software Development in the TestPoint 375 References 376 Chapter 14 Measurement, Control and Diagnostic Systems in Vehicles 377 14.1 Electronics in Vehilcles 377 14.2 Data Transfer Systems 381 14.2.1 Data Transfer Buses – Classification 381 14.2.2 CAN Bus System 382 14.2.3 FlexRay Transfer System 388 14.2.4 Media Oriented Systems Transport 395 14.3 Diagnostics 403 14.4 Sensors for Vehicles 406 Selected Bibligraphy 410 Acronomys and Abbreviations 411 About the Author 415 Index 417
- 13. 1 Chapter 1 Computer-Based Measurement Systems One of the components of measurement technology development is the devel- opment of measurement systems. By the measurement systems, we refer to a set of material and organizational resources, as well as software for information processing, aggregated in order to obtain, transmit, and process measuring data, and to display and store them. The measurement system is equipped with a personal computer (PC) or a microprocessor chip; its task is to control information flow in the system, to process measuring data, and sometimes to store them. The computer or the microprocessor chip is a system controller; that is a device managing the system. The measurement systems described in this book are exclusively digital systems. Measurement systems with the PC, called computer- based measurement systems, are of great importance. Considering the widespread use of PCs in both industrial and research measuring laboratories, the building of a computer measurement system usually does not imply the purchase of a separate computer, but allows the utilization of the existing ones. This is especially im- portant and economically effective in the case of building up computer measure- ment systems for the realization of temporary measuring tasks. Separate classes of computer measurement systems are the simplest two-component systems, composed of one measuring instrument plus one computer as the system controller. It is self-evident that the possibility of applying an existing computer for setting up the simplest measurement system decreases construction expenses of such a system considerably. 1.1 CONFIGURATION AND STRUCTURE OF MEASUREMENT SYSTEMS An important problem in designing and operating the measurement system is the organization of information flow in the system. Two criteria are essential for this organization:
- 14. Measurement Systems and Sensors 2 • The kind of transmission in the system: serial, bit-by-bit, or parallel, where the information is transmitted in the form of multibit words. According to this criterion, there are systems with serial interfaces and parallel interfaces. • The mode of information exchange between system devices with regard to the connection configuration of instruments: linear (bus), star, or daisy chain (arranged in rows). Measurement systems in the linear, star, or daisy chain configurations are shown in Figure 1.1. The linear configuration is used most often; in this configuration, the exchange of instructions passed between system devices is realized exclusively by the data bus of the system. The linear configuration is elastic because it makes change of the system structure readily possible by adding or disconnecting devices or by changing the placement of instruments in relation to other devices. Figure 1.1 Configuration of measurement system: (a) linear, (b) star, and (c) daisy chain. The star configuration requires the number of multibit computer inputs equal to the number of devices in the system (except for the computer). An advantage of this configuration is the fact that it does not address the bus devices because they are connected to determined computer inputs. Alteration of the structure of such a system is difficult, and sometimes impossible, since the measurement system con- tains a greater number of instruments. Even less elastic is the daisy chain configu- ration, in which the exchange is possible only between neighboring instruments. Such configuration is sometimes used in the case of simple measurement systems
- 15. Computer-Based Measurement Systems 3 with the only one way of information flow. In discussing measurement system configurations, it must be remembered that a lot of measurement systems consist of two components only: the controller and the measurement instrument. The problem of system configuration thus does not appear. The measurement system designed for measuring various physical quantities in the object consists of the following functional components: • A sensor or a set of sensors of physical quantities. The sensor causes changes of a definite electric parameter in the function of the value of a measured quantity (e.g., the resistance alteration in the function of tem- perature). • Measurement transducers, in which the electric parameter of the sensor is transformed into the direct voltage or the direct current (e.g., a transducer as the Wheatstone bridge with the bridge branch as resistance sensor with the electric voltage at the bridge output). • Conditioners or circuits standardizing the level of signal from the measure- ment transducer to the range of the input voltage of the analog-to-digital converter (ADC). • ADCs or digital measuring instruments containing such a converter; the task of ADCs is to convert analog into digital signals. • Devices for visual display of measurement results in the form of the display field of a digital measurement instrument, the screen of a digital instrument (e.g., the digital oscilloscope or the frequency analyzer), or a computer mo- nitor. • A computer with its software and memory resources. • Actuators or generators of test signals. • Power supplies of the object, operating autonomously or under control (op- tional). The flow-process diagram in a measurement system is shown in Figure 1.2. It is worth mentioning that the measurement system is often—particularly in in- dustry—part of a control-measurement system. Measuring data is for controlling objects measured, for diagnosing the objects, and monitoring their state. More complicated measurement systems can be constructed in the hierarchical structure. On the lowest level, there are measurement subsystems arranged to collect data from the object. Subsystems are situated on a separate area (e.g., in the manufacturing room or the laboratory). Data from such subsystems are sent to the main controller of the measurement system (see Figure 1.3). The main con- troller of the system not only receives initially processed measuring data, but it can also send commands relating to the execution of a measuring procedure or a set of commands for measurement instruments to subsystems. The main controller of the measurement system can also take advantage of memory resources, data display, and data storage devices.
- 16. Measurement Systems and Sensors 4 Figure 1.2 The flow-process diagram in a measurement system. These devices would not be effectively used in subsystems. The PC, Mac, laptop, tablet or similar class computer—as the main controller of the measurement system—can be programmed for a synthesis of collected measuring data and for advanced processing, as well as for data presentation. Interface systems on different levels of the hierarchical measurement system can belong to different interface standards. Figure 1.3 The hierarchical structure of the measurement system. In the example shown in Figure 1.3, each subsystem is composed in the stan- dard of the IEEE-488 parallel interface. All subsystems are united into the system by means of the local area network (LAN) computer network, the Ethernet type with serial transmission. For the implementation of such a configuration, the
- 17. Computer-Based Measurement Systems 5 computer in a subsystem must be equipped with the IEEE-488 interface board and the Ethernet network board. Some interface standards (e.g., Profibus or VXI) make it possible to build up hierarchical systems in the frame of one interface system. 1.2 INTERFACE SYSTEM A generally applied criterion of the division of measurement systems is a kind of transmission of digital announcements in the system (e.g., data, addresses, and commands); in other words, serial transmission or parallel transmission. The interface system assures equipment and programmatic adjustment of devices attached to the bus. According to the criterion mentioned, measurement systems are divided into the following categories: • Measurement systems with serial interface; • Measurement systems with parallel interface. The interface systems used most often in computer-based measurement systems are the following: the RS-232 serial interface and the IEEE-488 parallel interface (also called IEC-625, HPIB, or GPIB). 1.2.1 Interface System Meaning There is a notion of an interface system, as well as a notion of an interface only, which have a wider and a narrower sense, respectively. They are defined in standards. According to the standard, “interface is the coupling between a system considered and another system, or between devices of a system, through which information passes.” Interface in the narrower sense is only a matching-up circuit (e.g., matching-up signals of the TTL circuits to signals of the CMOS circuits), or adapting binary signals coded with voltage levels (e.g., logical 0 is 0V; logical 1 is 4.5V) for of binary signals coded with impulse frequency (e.g., logical 0 means 2,200 Hz; logical 1 means 1,200 Hz). The wider sense is given to the interface system, which according to the standard means, “the gathering of device- independent components—mechanical, electrical, and functional—necessary in the process of information exchange between devices.” Such gathering requires “cables, junctions, signal transmitters and signal receivers, interface functions with their logical description, the line signal, time relations as well as control rules.” Transmission protocols and control programs concerning system operation also belong to the interface system. In the common parlance, the notion of interface is often relevantly used instead of the notion of interface system. In this book, we will also take advantage of this abbreviation. It should be once again emphasized that the interface system describes (defines) the processing of only those signals that are transferred through an interface bus. Other signals in the
- 18. Measurement Systems and Sensors 6 measurement system, including very essential input measuring signals, both analog and digital, are neither defined nor standardized by the interface standard. 1.2.2 Interface Bus Signals transmitted through an interface bus bear the general name of interface messages. Interface messages are divided into data and instructions. The data transmitted is not only the result of measurement (measuring data), but also sets of instruments: measurement ranges, limit values for alarms, sets of power supply, sets of oscillators, the mode and the level of triggering oscilloscopes, and others. Instructions in the interface system are divided into commands and addresses. Certainly, the organization of the interface bus depends on the kind of interface. Parallel interface buses are more complex. The lines of the parallel interface bus are divided into groups that are also called buses. A separate bus of the parallel interface is always the data bus. The data bus contains 4 lines (Centronics) to 64 lines (PXI). The synchronization bus contains lines assuring time coordination between the sending and the accepting of data. The control bus (or interface management bus) contains lines destined for transmission of control signals. Control signals in a measurement system are the signal of resetting, interrupt request signals, commands of measurement execution, commands of generating a set of signals (for a generator in the system), and others. The address bus destiny is defined by its name. Binary addresses are transmitted across this bus. The addresses are sent to these devices, which ought to execute commands; related commands are available on the control bus. Quicker addressing takes place when the number of lines in the address bus is equal to the number of instruments included in the system. In such a case, addressing is performed with the “1 from n” method. In cassette (crate) systems with the parallel interface, a bus for clock pulses is set up. Sometimes, one part of the bus is the local rail, the line of which connects only neighboring modules in the cassette, and thus, as opposed to other lines in the interface bus, they are not led to all devices in the system. For particular systems of the parallel interface, the organization of the interface bus may differ considerably from the organization described above. For example, in the CAMAC system there are two separate data buses, each with 24 lines. One CAMAC data bus is provided for recorded data, the other for readout data. However, in the IEEE-488 system, the data bus serves not only for data transmission, but for addresses transmission as well. The bus of the serial interface can number two or more lines. The messages transmitted are organized according to careful rules and standards called communication protocols. The interface message frame contains both the receiver address and the data field; it also contains the field of control bits as well as redundant CRC bits for transmission validity check. The CAN and MicroLAN measuring-control systems have a similar data bus. In spite of a trivial opinion evaluating the number of lines of each serial interface to two, this bus can have anything from 2 to 35 lines, as in the RS-449 interface. They are mostly control
- 19. Computer-Based Measurement Systems 7 lines, but some of them also may be data lines (for received data, for transmitted data, or for the secondary channel). The full bus of the most popular RS-232C serial interface contains 22 lines. Only the elimination of some control and synchronization functions enables a decrease in the number of lines used in this interface to five (including the line of the ground), and, in the simplest version, even to two. 1.2.3 Interface Functions The following are interface functions used in linear configuration measurement systems. • Matching functions. Functions of adjusting consist of the processing of signals sent to the interface buses or received by the buses with a device into a standard form in the interface system. One can state that circuits acting in the interface system are interfaces in the narrower sense. • Synchronization functions. Synchronization is understood as the coordina- tion of data transmission in the function of time, realized in order to match the transmission speed of data to its reception speed. The type of transmission used most often in measurement systems is the asynchronous transmission with an acknowledgment of receipt in the so-called handshake mode. Another kind of transmission is the synchronous transmission. • Functions of buffering and error corrections. The correctness of the data transfer process in the system is checked. The fault detection of data received in a file often causes a requirement for the resending of the file or its fragment (frames) according to the procedure of the automatic repeating transmission of the determined data file, known as automatic repeat request (ARQ). It requires previous data storage in the buffer register. The buffering is also necessary when slight differences in the speed of data reception in relation to their transmission appear. In the case of unsynchronized transmission, the received data can be buffered in the receiver, and only after that readout. • Management functions. Functions of management consist of the control process of measurement and data processing, according to previously recorded programs and procedures. In particular, this function regulates the access to data transmitted by transmitters to the bus. It decides the sequence of events in confrontational events, and brings system devices to the initial state (the resetting). Devices included in the measurement system must have separate electronic circuits, enabling the realization of interface functions. Such devices are more complex than those not adapted for working in the system. This also pertains to computers working as controllers in measurement systems.
- 20. Measurement Systems and Sensors 8 1.3 MEASUREMENT ACCURACY AND MEASUREMENT SYSTEM DYNAMICS 1.3.1 Accuracy of Measurement Systems The accuracy of a measurement carried out with a measurement system depends on factors similar to those that determine the accuracy of measurement performed with a separate instrument that is not a part of a measurement system. Measurements performed with a measurement system are automated. In very general terms measurement can be defined as a comparison between the measured state Ax of a quantity and its reference state Aref, as shown schematically in Figure 1.4. Thus, the accuracy of measurement cannot be better than the accuracy of the standard. The accuracy of measurement or realization of a reference state is described by the measurement uncertainty. Figure 1.4 Measurement (schematic). The units of physical quantities and the standards for their realization are defined within the international system of units (SI). The SI system is founded on seven base units, which are the meter, kilogram, second, ampere, kelvin, candela, and mole. These are the units of seven base physical quantities. The derived units in the SI system are realized with the base units, the radian, and the steradian. The units of the seven base physical quantities are realized with particularly high accuracy. However, some derived units are realized with an even lower uncertainty (i.e., even higher accuracy) than the base units. The four units of physical quantities realized with the lowest uncertainty (i.e., with the highest accuracy) are the unit of frequency (realized by a cesium or rubidium atomic standard), the unit of length (realized with a laser interferometer), the unit of voltage (realized by a quantum standard based on the Josephson effect), and the unit of electrical resistance (realized by a quantum standard using the quantum Hall effect). The cesium standard for the realization of the units of frequency and time at the National Institute of Standards and Technology (NIST) in the United States allows the realization of the second with an uncertainty of 1016 . In many metrology laboratories the unit of length, the meter, is realized with an uncertainty of the order of 1014 . The quantum standard of the unit of voltage realizes the volt with an uncertainty of 109 , which is also the uncertainty of realization of the ohm, the unit of electrical resistance. A fundamental component of the uncertainty of any digital measurement is the quantization error. The relative quantization error quan is described by the formula:
- 21. Computer-Based Measurement Systems 9 quan = 1/Nx , (1.1) where Nx is the result of a measurement of a quantity A. The limit value of the absolute uncertainty A of measurement of a quantity A with a digital instrument, e.g. a digital voltmeter, is specified by the manufacturer of the instrument as a sum of two terms: A = read Ax + range Arange , (1.2) where Ad is the reading, Arange denotes the measurement range, read is the uncertainty term representing a percentage of the reading not smaller than the quantization error quan, and range is the uncertainty term representing a percentage of the measurement range. Example In voltage measurements a voltmeter is used with the relative uncertainty: read = 0.02% and range = 0.005% for the ranges of 10V and 100V. What is the absolute uncertanity V of voltage measurement when we measure the voltage of 3.0V (Vx = 3.0V)? When we used 10V-range of the voltmeter (Vrange = 10V) the absolute uncertainty is V = [(Vread)2 + (Vrange)2 ]0.5 = [(read Vx)2 + (range Vrange)2 ]0.5 = = [(0.02%3.0V)2 + (0.005%10V)2 ]0.5 = [(6 mV)2 + (5 mV)2 ]0.5 = 7.8 mV. When we used 100V-range of the voltmeter (Vrange = 100V): V = [(Vread)2 + (Vrange)2 ]0.5 = [(read Vx)2 + (range Vrange)2 ]0.5 = = [(0.02%3.0V)2 + (0.005%100V)2 ]0.5 = [(6 mV)2 + (50 mV)2 ]0.5 = 51 mV. Conclusion: In order to provide a small uncertainty the measuring range of an instrument should be chosen as low as possible. Manufacturers of instruments specify the uncertainty terms read and range in a table presenting the values of these terms for different measurement ranges and observation times. Usually the values of read and range are specified for the observation times of 24 hours, 90 days, and one year. When planning the measurement tasks to be executed by a measurement system we can have an influence on many factors related to the measurement control and the processing of measurement results, depicted schematically in Figure 1.5.
- 22. Measurement Systems and Sensors 10 Figure 1.5 Measurement of a quantity A with a measurement system. The measurement uncertainty can be improved by using procedures of automatic reset, automatic calibration, and statistical processing of results of a series of measurements. In metrology institutions (e.g., the Central Office of Measures in Warsaw, Poland, abbreviated GUM from the name Główny Urząd Miar) measurements with the lowest uncertainty are carried out with a measurement system. A single measurement carried out by a measurement system, in particular a system for measuring signals from multiple measurement points (e.g., many sensors) or for measurements of distributed objects, can have an accuracy worse than that of a measurement performed with instruments that are not part of a system. Such deterioration in the accuracy results from nonuniform characteristics of the sensors within the system and the increased risk of interference that can occur along with the measurement signal in a long measurement or transmission line to distort the measurement result. 1.3.2 Measurement System Dynamics The dynamics of automated measurements carried out with measurement systems is incomparably better than that of measurements performed with instruments that are not part of a system. The dynamics of a measurement system can be described by the duration of a single measurement cycle or by the measurement frequency (i.e., the number of measurements that can be carried out per unit of time). The highest requirements regarding the dynamics are imposed on real-time measurement systems (RTMS) and real-time measurement and control systems. A series of measurements with a period shorter than 0.1 second cannot be carried out otherwise than with a measurement system. As well, indirect (complex) measurements with a period of one second or shorter are only possible by means of a measurement system.
- 23. Computer-Based Measurement Systems 11 An indirect measurement involves simultaneous measurements of a number of physical quantities. For example, the electrical resistance R = f(V, I, T), which is a function of the voltage V, the current I, and the temperature T, can be determined by simultaneous direct measurements of V, I, and T. Figure 1.6 provides an example of system for such indirect measurement of nonlinear temperature-dependent resistance R over time. The resistance R is determined by simultaneous measurements of the voltage V across the terminals of the sample, the current I flowing through the sample, and its temperature T. On the basis of direct measurements of these three quantities the system controller calculates the value of R versus temperature. Figure 1.6 Indirect measurement of resistance R = f(V, I, T) with a measurement system. The dynamics of a measurement system depend on the following factors: • Conversion time of the analog-to-digital converter in the system for the assumed resolution of the analog-to-digital converter. • Transfer rate (bit rate) of the interface bus used by the measurement system. • Length of the interface bus; • Organization of transmitted messages, defined in the transmission protocol. Messages are created and transmitted in the form of communication (transmission) frames, which can vary widely in structure and size. The elements of a communication frame include fields of different length: address fields, control fields, a start frame delimiter and a frame check sequence field. For example, the CRC correction polynomial in the control field can have a length of 16 or 32 bits. • Algorithm of the computer program processing interface messages (data, instructions, addresses). • The number of measurement points at which simultaneous measurements are performed within the measurement system.
- 24. Measurement Systems and Sensors 12 • Performance of the system controller (i.e., the device (PC or microprocessor) controlling the operation of the system). In a distributed measurement system the controller can be a modern PC computer, typically a desktop computer or a notebook. In measurement systems with multiple components (measurement instruments, signal generators, recording devices) a desktop computer has the advantage of having more connectors and slots for connecting peripheral devices. The data processing speed of the computer does not limit the dynamics of the measurement system as a whole. Digital instruments with a short measurement time, autonomous fast analog-to-digital converters (see Chapter 6), or data acquisition (DAQ) cards with a high sampling frequency (see Chapter 13) can be used in a measurement system for measuring rapidly changing processes. Unless the system uses analog-to-digital converters with a long conversion time (e.g., integrating converters or delta-sigma modulators), the analog- to-digital conversion block does not limit the system dynamics either. The dynamics of a distributed measurement system is mainly determined by the time of transfer of messages via the interface bus and the properties of the software used by the measurement system. The software dependence of the measurement system dynamics is very complex and no general analytical approach can be used for its description. The only and fundamental rule in this case is that correctly developed software written in a higher-level programming language will slow down the operation of a measurement system as compared to software written in a lower-level language. Thus, the operation of a microprocessor system using software written in a machine language is potentially the fastest. Software written in an assembly language will be executed more slowly. Slower still will be the execution of programs written in general-purpose programming languages, such as Basic, Pascal, or C (C+, C++), depending on the complexity of the language used. By this rule, using software written in a development environment, such as LabVIEW or VEE, in a computer-based measurement system can result in the slowest operation of the system. Still, this only applies to comparison of the dynamics of operation of measurement systems that perform the same measurement procedure by the same operation algorithm. Measurements of the time of execution of instructions in a system were carried out at Poznań University of Technology, and the results show that, for some instructions can be executed faster in LabVIEW than in C++. This, however, is only possible because of the optimization procedures that the authors of LabVIEW implemented in this development environment. As illustrated by Figure 1.7, the time Taq of acquisition of a signal transmitted from a sensor to the computer used as the system controller is the sum of the time TA/D of conversion of the signal in an analog-to-digital converter, the time Ttr of transmission of the signal by a transmitter to the interface bus, and the time Tprop of propagation of the signal in the transmission line: Taq = TA/D + Ttr + Tprop . (1.3)
- 25. Computer-Based Measurement Systems 13 Measurement systems often use series transmission. The message transmission time Ttr is a function of the rate R of data transfer (data rate) by the interface bus and the size of the message, which is the size of the data frame or control frame, specified in bits: [bps] [bits] R n Ttr . (1.4) Figure 1.7 Message transmission in a measurement system and the system dynamics. The time Tprop of propagation of an electromagnetic (EMG) wave is a function of the velocity v of propagation of the wave in the cable (or in air) and the length l of the transmission line: Tprop = l/v. (1.5) If the length of the bus does not exceed 1 km, the time Tprop of propagation of an electromagnetic wave carrying a measurement signal is usually shorter than the transmission time Ttr. The role of the propagation time is sometimes underestimated in the calculation of the time balance of a system. We usually note that the velocity of propagation of electromagnetic waves, including light, in vacuum is 3 108 m/s, the upper limit of signal propagation speed in nature. With this value of propagation speed, an EMG wave will cover a distance of 1000 meters in a time of 3.3 µs. This, however, only applies to propagation in vacuum. In a transmission line in the form of an electrical cable or optical fiber the velocity of propagation of an electromagnetic wave depends on the quality of the cable or otical fiber, and is around 2 108 m/s. Transmission lines 1 km long are not unusual in measurement systems. In an electrical cable or optical fiber of this length the propagation time of an EMG wave will be about 5 µs. Expressed in samples per second (sps), the sampling frequency S (i.e., number of samples transmitted to a computer per unit of time), depends on the acquisition time Taq and can`t be greater than 1/Taq: S 1/Taq . (1.6)
- 26. Measurement Systems and Sensors 14 For example, let us calculate the acquisition time Taq and the maximal sampling frequency S in a measurement system with the analog-to-digital conversion time TA/D = 100 s, the data frame size n = 100 bytes = 800 bits, the interface bus data rate R = 1 Mbps, and a transmission line of length l = 1 km: TA/D = 100 s, μs 800 Mbps 1 bits 800 R n Ttr , μs. 5 m/s 10 2 m 1000 8 prop T The data acquisition time Taq in this measurement system will be Taq = TA/D + Ttr + Tprop = 905 s. The sampling frequency S, i.e., the number of samples processed by this system per unit of time, will be: S 1/Taq = 1/905 s = 1100 Sps. A message frame transmitted in a different interface system can be much longer than 100 bytes, the value provided in the example above. For instance, in the Ethernet or Fast Ethernet standards for local area networks (LANs) the length of a communication frame must be in the range from 64 bytes to 1518 bytes (see Chapter 12). It is worthy of notice that an acknowledgment of receipt of transmitted data is requested in many measurement systems (and transmission systems in general). An acknowledgment of receipt of data contained in a communication frame is sent by a computer to the measurement station. Thus, the time necessary for the computer to send the acknowledgment message and its propagation time must also be taken into account in the data acquisition cycle. A communication frame with an acknowledgment, with a typical length of the order of tens of bits, is usually much shorter than a communication frame with measurement data. Thus, the total time of transfer of the data and the acknowledgement of their receipt is 10% to 20% longer than the acquisition time Taq. However, it is noteworthy that some networks used as a measurement system bus (e.g. Et,hernet or Bluetooth networks) have asymmetric data rates. In this case, the uplink data rate (i.e., the rate at which data are sent from a computer) is much lower than the downlink data rate (i.e., the rate of data transfer to the computer) (see Chapters 8 and 12). The above-presented evaluation of system dynamics does not take account of the effect of interference and the related necessity of repeating the transmission of a communication frame or data packet many times. In an electrical transmission line interference occurs due to induction of undesired signals by the external electromagnetic field, as a result of transient states in the transmission of pulse signals, and due to long-line wave effects in the cables. The level of interference can depend on the rate of transfer of a sequence of bits (i.e., on the signal frequency). Interference and the related problems in measurement systems are discussed in the next section. The transmission protocol of a measurement system often includes the
- 27. Computer-Based Measurement Systems 15 ARQ function, which forces retransmission if an error is detected in a received frame or packet. Obviously, the necessity of retransmission of a part of messages slows down the operation of the whole system. Sometimes the system dynamics is specified synthetically for a single measurement point (e.g., as one hundred measurements per second). When measurements are carried out at a number of measurement points the dynamics of the system is worse. For example, a system with five measurement points will perform twenty measurements per second per measurement. 1.4 INTERFERENCE PROTECTION 1.4.1 Interference in Measurement Instruments Interference reduction is a very important issue in industrial measurement systems as well as in many other measurement setups. Interference in a measurement system is disturbance of the useful signal due to undesired signals that occur along with it. Interference can reduce the range of measured signals, distort the acquired information, or even damage components of the system. Sources of interference include external objects (e.g., radio stations, electrical motors, or atmospheric discharges), signals within the system (e.g. a difference in the ground potential between distant system devices), as well as phenomena that distort the intended signal, such as transient states in the transmission of pulse signals, wave effects in a transmission line, or undesired wave propagation in a transmission channel. The intended signal can be distorted also by the inherent noise of electronic components and elements; however, the underlying causes of noise are fundamentally different from those of the other types of interference. In general, noise occurs due to the granular structure of matter. It tends to have a lower power than other interference signals. Noise and its reduction are not discussed further in this book. Three elements in the course of interference must be considered when addressing the problem of interference and its effect on a measurement system or a single measurement instrument: the source of interference, the coupling channel, and the receiver of interference, as illustrated by Figure 1.8. Figure 1.8 Course of interference in a measurement system or instrument. Interference enters a measurement system through the following types of coupling circuits:
- 28. Measurement Systems and Sensors 16 • Impedance circuits. • Common resistance circuits. • Electromagnetic coupling circuits. The schematic diagram in Figure 1.9 shows how a source of interference acts on a signal receiver via an impedance coupling circuit. Figure 1.9 Interference acting on a signal receiver via an impedance coupling circuit. A measurement signal (or, in general, any useful signal) V1 is brought to the signal receiver by conductors, which can be wires or conductive tracks on a printed circuit board (PCB); the wires or conductive tracks have a resistance Rc. Along with the intended signal, an interference voltage signal V2 also reaches the receiver, since the impedance Zi of the isolation between the receiver and the source of interference is not infinitely large, but finite. The source of voltage V2 can be of use or even indispensable in the system, for example as a generator of synchronization signals. However, for an analog receiver it is a source of interference. Thus, the voltage Vin at the input of the receiver is a function of both the measurement signal voltage V1 and the interference voltage V2. The resistance Rp of the wires or PCB conductive tracks is negligible with respect to the input resistance Rin of the receiver. The wire resistance Rp is of the order of 100 m, while the receiver input resistance Rin can be of the order of 1 M. The impedance of the isolation is difficult to estimate, but can be assumed to be of the order of 100 M. The impedance decreases with increasing frequency of the signals, and the effect of interference on the input voltage Vin grows with frequency. Generation of interference in a coupling circuit sharing a common resistance of the power supply circuit is shown in Figure 1.10.
- 29. Computer-Based Measurement Systems 17 Figure 1.10 Interference acting on a signal receiver through a common resistance of the power supply circuit. In this diagram two circuits in a system or instrument use the same power supply. The power supply circuit generates a voltage Vp. The resistance of wires or conductive tracks on a printed circuit board is denoted R in the diagram. In the case of conductive tracks the resistance can be significant. Conductive tracks are made of copper sheet of a standard thickness of 70 µm. Thus, for example, a 100 mm long, 5 mm wide, and 70 µm thick copper track will have a resistance R = 5 m (with a copper resistivity value = 17 109 ∙m assumed in the calculation). The supply voltage may vary widely if one of the circuits draws a high and time-variable supply current. Such pulses of power supply current consumption are characteristic of digital circuits, in particular those with synchronous operation coordinated by a clock signal. When switching between logic states digital circuits of all types draw a power supply current many times larger (TTL circuits) or many orders of magnitude larger (CMOS circuits) than the current in the static state. Thus, in a synchronous digital circuit the current drawn by many components during switching is much higher than the mean supply current. In the diagram shown in Figure 1.10 the circuit 1 is a digital circuit drawing supply current in pulses from I1 to (I1 + I1), and the circuit 2 is an analog circuit drawing a constant supply current I2. Thus, the supply voltage V1 is described by the equation (1.7) in a static state and by the equation (1.8) during switching of digital components in the circuit 1: V1 = Vp 2R(I1 + I2) (1.7) V1 V1 = Vp 2R(I1 + I1 + I2). (1.8) The change V1 = 2R∙I1 in the supply voltage does not affect the operation of the digital circuit as long as it remains within the power supply tolerance of the digital circuit (e.g., for TTL circuits Vp = 5 V 0.25 V). The supply voltage V2 of the analog circuit 2 in Figure 1.10 changes in the same manner as the voltage V1, which means that V2 = V1. Changes in the supply voltage V2 result from changes in the current consumption of the digital circuit 1:
- 30. Measurement Systems and Sensors 18 V2 = Vp 2R(I1 + 2I2) (1.9) V2 V2 = Vp 2R(I1 + I1 + 2I2) = V2 V1. (1.10) In contrast to digital circuits, which are insensitive to slight fluctuations of the supply voltage, the output signal of analog circuits strongly depends on the supply voltage. The ratio of the interference signal Vint to the output voltage signal Vout can be estimated as equal to the ratio of the change in the supply voltage to the nominal supply voltage: Vint/Vout = V2/V2. (1.11) An efficient way to prevent the above-discussed interference consist in full separation of the analog circuits from the digital circuits, or at least the use of separate reference potential (ground) wires or tracks for the analog and digital circuits. A measurement instrument, in particular a high-sensitivity instrument, can be sensitive to the surrounding electromagnetic field. To reduce the effect of interference related to this field measurement instruments are often enclosed in a metal frame, or chassis, which plays the role of a shield (screen) protecting the instrument against electromagnetic field, or only against electric field in the case of shields of nonmagnetic materials, such as aluminum. High-sensitivity measurement instruments are always enclosed in a metal frame. A measurement instrument (e.g., a multimeter) in a metal chassis, which also plays the role of electromagnetic shield, has four terminals on the front panel: two input terminals labeled HI (for “high level”) and LO (for “low level”), a ground terminal labeled , and a shield terminal S (see Figure 1.11). Alternatively, the shield and ground terminals can be, and often are, mounted on the rear panel of the instrument rather than on the front panel. The earth ground potential is provided by a separate power supply wire of a typical three-wire power cable used in electrical grids. Now we consider three terminals of an input circuit of the instrument: signal terminal LO (low), shield terminal S, and ground terminal . All users of measurement instruments know that two signal wires (the transmission line) should be connected to the HI and LO input terminals. However few users only know what to do with the shield terminal S. Usually the shield terminal S is left unconnected, or is shorted with the ground terminal , or connected to the LO input terminal. Before elucidating the role of the ground and shield terminals let us discuss the input circuit of a measurement instrument, depicted schematically in Figure 1.11.
- 31. Computer-Based Measurement Systems 19 Figure 1.11 Shielded measurement instrument with four input terminals: signal terminal HI (high), In Figure 1.11 Rr1 denotes the resistance between the HI terminal and the virtual ground (point VG) of the measurement instrument; Rr2 is the resistance between the LO terminal and the virtual ground VG. The sum Rr1 + Rr2 = Rin. The values of Rr1 and Rr2 are of the order of M. The values of Rg1 and Rg2, the resistances between the virtual ground, and the shield, or the ground point, respectively, are lower than 1 . The resistances RL1 and RL2 of the wires of the transmission line depend of the type, quality, and length of the cable, but can be assumed to be in the range from 0.1 to 10 . A general rule to be obeyed in order to reduce the effect of interference in a shielded measurement system is: The shield S must always be connected so that no common-mode current flows thought the input resistances. If the shield terminal S of the measurement instrument is left unconnected, an interference current caused by the common voltage source Vsum and the ground potential difference Ve will flow through the input circuit resistances RL2, Rr2, Rg1 and Rg2. This is an adverse configuration of the circuit. As well when the shield terminal S is short-circuited with the ground (earth) terminal , the two sources of interference voltage (Vsum and Ve) will produce an adverse effect. In this case an interference current will flow through three input circuit resistances, RL2, Rr2, and Rg1. The best interference protection is ensured by connecting the shield terminal S with the terminal of the source of the measured voltage Vx (see Figure 1.12).
- 32. Measurement Systems and Sensors 20 Figure 1.12 Shielded measurement instrument with a correctly connected shield terminal S. In Figure 1.12 the sources of interference signals Vsum and Ve are shorted to the chassis shield and do not affect the input circuit of the measurement instrument. 1.4.2 Interference Induced in Transmission Lines Interference in measurement systems is also induced in transmission lines. The level of interference induced in a transmission line depends on the following factors: • Magnitude of the electromagnetic field in the surroundings of the transmission line. • Length of the transmission line (cable). • Type of cable used in the transmission line. • Terminal impedance or terminal resistors connected to the end of the transmission line or bus. By Faraday’s law of induction, the magnitude of the electromagnetic field and the length of the cable exert an influence on the voltage induced in the line. The cable plays the role of antenna for interference, and the type of cable is of much importance for the interference level. The lowest interference voltage is induced in a shielded, particularly coaxial cable. Coaxial cabling is sometimes used in the wiring of measurement systems; however this type of cable and its installation is expensive. Moreover, coaxial cabling requires expensive specialist coaxial sockets, plugs, and multiplug sockets. Twisted pair cabling is a cheaper solution for a transmission line. A variety of twisted pair, unshielded twisted pair is the type of cable the most commonly used in distributed measurement systems. The simplest and cheapest type of cable, untwisted pair is a much better antenna for interference,
- 33. Computer-Based Measurement Systems 21 and for this reason it is rarely used in the wiring of measurement systems because unfavorable for interference reduction. However, untwisted pair cabling is sometimes used for transmission of measurement signals. For example, electrical wiring within a power grid can be used for transfer of measurement and control messages (power line communication, PLC). This solution is discussed in Section 7.7. Designers and users of measurement systems rarely have influence on the source of interference and the power level of signals it generates. By contrast, much can be done in this regard with the receiver of interference and the coupling circuits. A receiver of measurement signals is also a receiver of interference. Differential Input Circuit The effect of interference can be substantially reduced by using differential transmission circuits, rather than asymmetric circuits, in the transmission line, as shown in Figure 1.13. Figure 1.13 Transmission in a measurement system with interference affecting the transmission line: (a) asymmetric transmission circuit; (b) differential transmission circuit. In a system with an asymmetric transmission circuit (Figure 1.13a) a measurement signal V1 (analog or digital) from the transmitter reaches the receiver along with an interference signal induced in the signal wire (called the “hot” wire) of the transmission line. By contrast, in a system with a differential transmission circuit (Figure 1.13b) the transmitted information is contained in the potential difference (V1 V2). If the same interference voltage is induced in both wires of the transmission line, the potential difference at the input of the receiver is the same as the potential difference at the output of the transmitter in this system. The interference voltage signal Vint is a common signal (in-phase signal) of both input terminals of the receiver.
- 34. Measurement Systems and Sensors 22 Transmitter output (differential circuit): V = V1 V2 Receiver input (differential circuit): V = (V1 + Vint) (V2 + Vint) = V1 V2. A differential amplifier in the receiver will amplify the potential difference at the inputs of the receiver and attenuate the common (in-phase) input signal, including the interference signal. Differential amplifiers are discussed in Chapter 5. Note that in Figure 1.13 the ground potential of the transmitter and receiver have different symbols. This should remind, that distributed (distant) electrical devices can have different ground potentials. In industrial installations (e.g., in the metallurgical industry) differences in the ground potential at distant points can even exceed 10V. Wave Effects in Transmission Line If the length of the electric wires in a transmission line is of tens of meters or more, long-line wave effects will occur in the line. Moreover, in a transmission line of such length the signal will be substantially attenuated (similarly to a radio signal; see Chapter 8). The attenuation effect depends on the path of the signal, its frequency, and the quality of the cable, and can be easily determined and corrected. If the attenuation is strong, the transmitted signal should have a sufficiently large amplitude or should be amplified either on its way or in the receiver. The influence of wave effects is more difficult to determine, but there are simple methods that allow to reduce it. A digital signal is generated and transmitted in the form of pulses of voltage or current or as a sine signal with modulated frequency. A pulse signal is conveyed by amplitude-shift keying (ASK) or quadrature amplitude modulation (QAM). Frequency-shift keying (FSK) is used in the case of sine signal. Steep voltage pulses distort the signal waveform by causing two effects: • Overoscillation and overvoltage due to transient states that occur as a result of unit-step forcing in an RLC transmission line; • Signal reflection in a line shorted or open at the end. The risk of overvoltage is the reason for which the RS-232C interface standard, discussed in Chapter 7, sets a limit to the pulse slope (i.e., the rate of change of signal voltage) at 30 V/s. The steeper the voltage pulses, the larger the signal distortion and overvoltage. A transmission line has characteristics of a long line if its length is comparable with the wavelength of electromagnetic waves transmitted by this line. A sequence 1/0/1/0 of digital signals transmitted at a rate of 10 Mbps can be regarded as a square signal with a frequency fs = 5 MHz. The velocity v of an electromagnetic wave of this frequency traveling in an electric cable is 2 108 m/s, approximately; thus, the wavelength of the electromagnetic wave is = v/fs = 40m. Wave effects,
- 35. Computer-Based Measurement Systems 23 including reflection, can be observed in segments of the transmission line of a length of /4, /2, or more. Figure 1.14 shows traces of signals in a 50m long transmission line made with coaxial cable. The measurement system in which these signals were observed and measured is shown in Figure 1.15. Figure 1.14 Transmission of pulse (digital) signal over a long line. Panels a, b, c, and d in Figure 1.14 present the output signal of the generator, the signal at the input of the line closed with a 50 resistor, the signal at the input of the line open at the end, and the signal at the input of the line shorted at the end, respectively. Note that the closing of the long line with a 50 resistance (a value close to the wave impedance of the line) in Figure 1.14b eliminates the effect of signal reflection. The recorded signal only has a smaller amplitude compared to the generator output signal. Shown in Figure 1.14c, the signal at the input of the line open at the end results from the superposition of the generator output signal and the signal reflected at the end of the line. The reflected signal returns to the input of the line in the same phase, but is shifted in time due to the propagation delay.
- 36. Measurement Systems and Sensors 24 Figure 1.15 System for measurement of wave effects in a transmission line with characteristics of a long line. In Figure 1.14d the signal at the input of the line shorted at the end results from the subtraction of the signal reflected at the end of the line from the input signal. The reflected signal returns to the input in the opposite phase and is shifted in time due to the propagation delay. The plots involving the reflected signal (Figure 1.14c and d) for the open and shorted line show distortion in the form of steps that result from the propagation delay and the superposition or subtraction of the reflected signal. The discussed signal reflection in a long line can be easily prevented by closing the transmission line with a 50 to 150 terminal resistor. Energetically matched to the line, such a resistor absorbs the signal energy at the end of the line, and thus prevents signal reflection. The matching to the end of the line is very important in signal lines in measurement systems, since signal distortion can substantially degrade the reading of the digital information carried by the signal. Selected Bibliography Boyes, W., (ed.), Instrumentation Reference Book, 3rd ed., Boston, MA: Butterworth-Heinemann, 2003. Garrett, P. H., Advanced Instrumentation and Computer I/O Design: Real-Time System Computer Interface Engineering, New York: IEEE Press, 1994. Nawrocki, W., Computer-Based Measurement Systems, Warsaw, Poland: Wydawnictwa Komunikacji i Łączności, 2nd ed., 2006 (in Polish). Ott, H.W., Noise Reduction Techniques in Electronic Systems, Wiley, 1988. Park, J., and S. Mackey, Practical Data Acquisition for Instrumentation and Control Systems, Amsterdam, the Netherlands: Elsevier, 2003.
- 37. 25 Chapter 2 Computers for Measurement Systems 2.1 FUNCTIONS OF COMPUTER IN MEASUREMENT SYSTEM The increasing significance of computer-based measurement systems results from a widespread use of computers and software. The PC in standard configuration may operate as a controller of simple measurement systems. Such possibilities of controlling a measurement system is given by junctions and drivers of the universal serial bus (USB), the RS-232 interface, as well as of the PCI and PCI Express slots, installed conventionally. Controlling measurement system with other interfaces requires the addition of an additional board or controller module to the computer. The functions of the PC in the measurement system are as follows: • Functions of controlling the system (the controller of the measurement system); • Functions of data processing in wide range, using computer programs such as Matlab or Excel; • Functions of servicing peripheral devices, such as instruments (remote control), monitor, and keyboard; • Data storage; • Controlling data transmission outside the measurement (e.g., via the Internet). In the case of the measurement system without a computer (e.g., the system with a microprocessor), a majority of the functions mentioned must be also fulfilled in the system, sometimes only in the narrower range. In particular, the installation of peripheral devices such as a keyboard and a monitor, as well as data storage, are essential in every measurement system, and data processing is usually very advisable.
- 38. Measurement Systems and Sensors 26 2.2 TYPES OF COMPUTERS FOR MEASUREMENT SYSTEMS All types of computers are used in measurement systems, with the exception of supercomputers, which have the highest computational capacity. Specifically, computers used in measurement systems include: • Desktop PCs. • Laptops. • Tablets. • Compute modules (e.g., Raspberry Pi). • Smartphones and phablets. The most commonly used of these are desktop PCs. A desktop PC plays the role of a control device in a fixed measurement system. Mobile or portable measurement systems use laptops instead. As well, inexpensive tablets or smartphones (or phablets) are used in distributed measurement systems as controllers installed in distant measurement stations. A desktop PC in a measurement system has three important advantages over a laptop or tablet: • Higher efficiency, resulting, among other factors, from a better processor, which can operate faster in a desktop PC because of a higher power supply with better cooling conditions. Good cooling is ensured by a larger and more efficient ventilator, larger radiators, and a larger space around the processor for heat exchange. • Larger number of installed peripheral connections. The connectors that are of use in fast-operation systems include internal connectors (i.e., slots installed on the motherboard colloquially known as the mobo). Fast- processing data acquisition (DAQ) boards and IEEE-488 (GPIB) interface cards can be connected to PCI or PCI Express slots inside a desktop PC. • More possibilities for maintenance and expansion. In contrast to a desktop PC, which has multiple communication links, a tablet, smartphone, or phablet usually only has four links of potential use for connection to the interface bus for a measurement system. These include three radio links, WiFi, Bluetooth and a GSM modem (optional in tablets), and a micro- USB wired connection (some tablets have two USB connectors). Table 2.1 provides the parameters of computers typically used in measurement systems. Table 2.1 does not include data on external or internal slots for connecting a monitor or projector (HDMI connector), audio devices (minijack for headphones, loudspeakers, microphone), or memory chips and cards (such as microSD). Apart from the computers mentioned above, simple and inexpensive (costing $100) compute modules are also available on the market. Although they are similar in appearance and size to boards with microprocessor systems,
- 39. Computer-Based Measurement Systems 27 compute modules are classified as computers because of their architecture and much higher efficiency and functionality. Table 2.1 Technical Data of Computers Used in Measurement Systems (Ranges of Typical Parameters) Parameter PC Laptop Tablet Processor clock rate 2 GHz to 4 GHz 1 GHz to 4 GHz 266 MHz to 1 GHz RAM 2 GB to 32 GB 1 GB to 32 GB 1 GB to 2 GB Drive 1 TB to 2TB 512 MB to 2 TB 16 GB to 132 GB flash Peripheral interfaces (standard) USB 2.0, USB 3.0 Ethernet COM USB 2.0, USB 3.0 Ethernet 10/100, WiFi (IEEE 802.11n, IEEE 802.11ac) Bluetooth, IrDA USB 2.0, USB 3.0 Bluetooth, WiFi (IEEE 802.11n, IEEE 802.11ac) Peripheral interfaces (optional) RS-232C, IEEE 1394 WiFi (IEEE 802.11n, ac) Ethernet 1000, IrDA, Bluetooth RS-232C IEEE 1394 GSM modem GSM modem . Figure 2.1 Raspberry compute module. The Raspberry Pi compute module has become quite popular in recent years and has been also successfully used as a measurement system controller. A central component of Raspberry Pi is a BCM2835 processor (Broadcom, USA) with a rather high computational capacity, which allows the use of an operating system such as Linux or Android. The BCM2835 processor has a 700-MHz clock and
- 40. Measurement Systems and Sensors 28 512-MB RAM. The compute module has a slot for an SD memory card with an operating system. The SD memory card is also used for data storage. Raspberry PI has also two USB connectors, an HDMI connector (usually used for connecting a monitor), and an RJ45 connector for Ethernet. 2.3 COMPUTER ARCHITECTURE The modular structure of modern PCs makes it possible to connect additional circuits (e.g., boards, memory circuits), as well as peripheral devices to various points of computer buses. The architecture of PCs has changed over time from the multibus architecture [Figure 2.2(a)] used from 1992 to 2007 to the current architecture [Figure 2.2(b)] used in computers manufactured since 2007. The central component in the current architecture is North Chipset (Chipset 1), controlled from the processor and controlling many buses and peripheral devices. The point of joining the interface board to the computer is very important with regard to the data processing rate. The further from the processor (in the computer architecture) the interface board is, the slower the communication with this board and the operation of the board itself will be. In Figure 2.2 a simplified block diagram of the PC motherboard and the distribution of buses in the computer are shown along with the arrangement of junctions provided by constructors for joining additional or external devices. In Figure 2.2(a) and 2.2(b), neither buses nor nonessential ports for the measurement system are shown, nor are the E-IDE bus (for hard disks and CD), the SCSI bus, PS2 junctions for a keyboard or a mouse, or junctions for a floppy disk. Additional computer boards are joined to sockets inside the computer in the space provided for this purpose, in the so-called slots. External devices are joined to junctions (sockets or connectors, according to the type of interface) mounted into the computer casing. The multibus architecture [Figure 2.2(a)] was used in computers manufactured until around 2007. Computers with this architecture are still used in many industrial and research measurement systems. The architecture shown in Figure 2.2(a) has three buses: a Front Side Bus (FSB) for data exchange between the processor and both the random access memory (RAM) and the cache memory, a high data rate Peripheral Component Interconnect (PCI) bus, and a slow data rate Industry Standard Architecture (ISA) bus.
- 41. Computer-Based Measurement Systems 29 Figure 2.2 Buses and their junctions in a desktop PC: (a) the architecture of PC manufactured before 2007 and (b) the architecture of PC manufactured after about 2007.
- 42. Measurement Systems and Sensors 30 The most essential integrated circuit of the computers is the processor. The processor is connected to the remaining part of the computer with the FSB bus. Probably the fastest microprocessors for the PC in current use at the time of this writing (2015) are the Intel Core i7- 4770K and A10-7850K manufactured by Intel and AMD, respectively. They have a clock frequency of 3.2 GHz (in the turbo mode the frequency can reach 5 GHz). Connected to the FSB rail, the integrated circuit north chipset (Chipset 1) is the controller of RAM and cache memory. Chipset 1 is the interconnector between the FSB bus and a basic bus of the computer, the PCI bus. Cache is a very fast memory of the processor, which serves to store the most often used instructions and data, applied in order to accelerate the information processing. The south chipset integrated circuit (Chipset 2) on the motherboard of the computer goes between the PC bus and the ISA bus—the second most important bus in the computer PC. A very important circuit in the computer structure is the input/output controller. The circuit of the input/output controller contains drivers of these interface systems, the junctions of which are placed on the computer casing: the RS-232 interface serial. Table 2.2 Buses in a PC and in a Laptop Bus Number of Bits Transmission Rate (maximum) Notes PCI 32 64 132 MBps (at 33 MHz) 512 MBps (at 66 MHz) Parallel, for PC boards PCIe × 1 1 250 MBps Serial, for PC boards PCIe × 4 4 1 GBps Parallel, for PC boards PCIe × 8 8 2 GBps Parallel, for PC boards PCIe × 16 16 4 GBps (version 1.0) 8 GBps (v2.0) 16 GBps (v3.0) Parallel, for PC graphics card USB 2.0 USB 3.0 USB 3.1 1 480 Mbps 5 Gbps 10 Gbps Serial, for peripherals IEEE-1394b 1 800 Mbps Serial, for peripherals Ethernet 1 10 Mbps 100 Gbps Serial, for LAN IEEE 802.11n 1 150 Mbps Carrier frequency 2.4 GHz IEEE 802.11ac 1 7 Gbps With multiple antennas, carrier frequency 5.8 GHz Bps = bytes per second; bps = bits per second.
- 43. Computer-Based Measurement Systems 31 The PCI bus is equipped with a 32-bit data bus controlled by a clock signal with a frequency of 33 MHz (in the PCI 2.1 version, a frequency of 66 MHz), that provides the data rate of 132 MBps (megabytes per second), and in the case of the 66-MHz clock, the rate of 264 MBps. However, high rates can be obtained only in serial mode of transmission (burst mode), which assumes that a single addressing of the data receiver is followed by the transmission of a data block with any volume. The recording of the full word (32 bits), the so-called single write, which requires at least two clock cycles, or the reading of the full word (that is to say; “single read”), is slower: it requires three clock cycles. The transmission of full words goes on with the top rate of 44 or 66 MBps. The data bus of the PCI bus can be extended to 64 bits. Basic parameters of the buses discussed are introduced in Table 2.2. In the computer architecture, the PCI bus is located nearest to the processor, and, therefore, the transmission rate on this bus is higher. In this estimation, the FSB bus is omitted, to which no additional devices are joined, except for addi- tional RAM memory circuits. Three or two slots for PCI devices are set up onto the motherboard of the computer. Since 2007 the computer architecture has been as shown in Figure 2.1b. In modern computers the memory controller is included in the processor. North chipset has individual connections for drivers of peripheral devices compatible with the Peripheral Component Interconnect Express (PCIe) bus. South chipset is an integrated peripheral device driver. Modern computers retain the PCI bus to allow connection of older modules compatible with the PCI standard. A motherboard of a PC with slots for PCI and PICe devices is shown in Figure 2.3. Computer boards serving devices of high operation speed are connected to the PCI bus, including an interface board, measuring boards, and input/output (I/O) boards. For example, in measurement tasks it is possible to join the following types of computer boards manufactured by National Instruments to the PCI bus: • LAN board, (mostly Ethernet board); • PCI-GPIB board of the parallel IEEE-488 interface controller; • PCI-485 controller board of the RS-485 serial interface with two to eight ports; • PCI-CAN controller board of the measurement system with the CAN serial bus; • NI 5911 board of the analog-to-digital converter (I/O board); • PCI-DIO-24 Data AcQuistion Board (DAQ), the multichannel I/O board; • Oscilloscope board, serving to set up the virtual oscilloscope; • PCI-1422 board, 16-bit board of the ADC for visual signals.
- 44. Measurement Systems and Sensors 32 Figure 2.3 A motherboard of a PC with slots for PCI boards (two long white connectors) for a PCIe ×16 board (long black connector) and for a PCIe ×1 board (short white connector). Computer boards of other measuring devices manufacturers, like Keithley, are also designed to be connected to the PCI bus. After 2000 in the measuring technology modular measurement systems, PCI extensions for instrumentation (PXI) with PCI bus as an interface bus are used. The PCI bus is installed in both PCs and power-PCs (e.g., manufactured by Apple), as well as in workstations. Thus, the interface board or the PCI measuring board can be composed into a computer measurement system with computers of these classes. The ISA bus has a 16-bit data bus (in older versions an 8-bit bus), controlled by the clock with a frequency of 8 MHz. Because the data transfer on the ISA bus requires at least two clock cycles (and in many cases even eight cycles), the top rate of data transmission with the ISA bus amounts to 8 MBps, or, in other words, it is 32 times lower than the transmission rate on the PCI bus carried on in the serial mode (burst). Therefore the top speed operation of measurement systems using the ISA bus is lower than systems with the PCI bus. Various types of computer boards designed for measurement systems can be connected to the ISA bus: • AT-GPIB (by NI) or KPC-488 (by Keithley), the IEEE-488 interface controller board; • AT-485 (by NI), the RS-485 interface controller with 2 to 8 ports; • AT-232 (by NI), the RS-232 interface controller with 2 to 16 additional ports (the PC computer conventionally has two RS-232 ports installed);
- 45. Computer-Based Measurement Systems 33 • AT-CAN (by NI), the controller board of the measurement system with the CAN bus with 1 or 2 ports; • The I/O boards contain an analog-to-digital converter and a digital-to- analog converter [e.g., NI 5102 (by NI)]. The chipset 2 circuit controls the USB, with a transmission rate up to 480 Mbps (USB 2.0). In these computer measurement systems in which a laptop is the system controller, parameters of the Personal Computer Memory Card International Association (PCMCIA) bus are essential. The PCMCIA bus is set up in computers of this class and a similar PCI bus in PCs. The PCMCIA ports were introduced in 1989 in order to connect additional memory boards to the portable computer. Instead of the last Version 2.0 of the PCMCIA standard, the Card Bus standard was introduced in 1994, with parameters similar to the PCI bus. Essential differences between the PCMCIA Version 2.0 and the Card Bus consist of the extension of data wordlengths from 16 bits (PCMCIA) to 32 bits, and in increasing clock frequency of the bus up to 33 MHz. In the new standard (Card Bus), the dimension of the PCMCIA boards and 68-pin junctions to the bus are retained. The PCMCIA name is still applied for both the bus and junctions and for computer boards fulfilling the conditions of the Card Bus standard. The power supply voltage of PCMCIA boards is 3.3V. The PCMCIA junction serves to join additional memory cards or additional hard disks to the laptop. Moreover, for such junctions, one can connect the following measuring devices in the form of the PCMCIA boards: the interface board (IEEE- 488, RS-232C, CAN), the DAQ (measuring board) with ADCs and DACs, the modem board of the PSTN telephony, and the modem of the GSM cellular telephony. The PCMCIA boards have the following standard dimensions: a length of 85.6 mm, a width of 54 mm, and a thickness d different for three card types: the PCMCIA type I—the thickness d = 3.3 mm; the PCMCIA type II—d = 5 mm; and the PCMCIA type III—d = 10.5 mm. At present, in modern laptops slots for PCMCIA boards are seldom installed. 2.4 UNIVERSAL SERIAL BUS Measurement systems are usually very simple systems composed of one digital instrument and one computer (see Chapter 7). For construction of such a system, one can use the RS-232C serial interface installed in almost every PC. An essential limitation of the system with this interface is a transmission rate not higher than 20 kbps for transmission lines with a length of 15m, and not higher than 115 kbps for systems with a short transmission line of 1.5m. Data transmission in the computer system and in the computer-based measurement system can be realized with a rate considerably higher than in the RS-232C interface.
- 46. Measurement Systems and Sensors 34 The USB and the IEEE-1394 (FireWire) create a new method of attaching and accessing peripheral devices, which simplifies the attachment and configuration from the end-user point of view. The USB, manufactured since 1997, allows a transmission rate up to 10 Gbps (USB 3.1) and up to 480 Mbps in USB 2.0 version (widely applied). The USB belongs to the PC structure or to the laptop structure. Still higher transmission rates than USB 2.0 up to 800 Mbps can be obtained by using the IEEE-1394 serial bus. For several years, the IEEE-1394 bus has been installed in computers, both in the physical layer and in the form of bus drivers in the Windows operating system. As far as the usage in the measurement system is concerned, there is a qualitative difference between the RS-232C serial interface and the IEEE-1394 serial bus, or the USB serial bus. The difference is that the RS-232C interface system is simpler, with regard to electric and mechanical parameters, organizational rules, and transmission protocols. The USB or the IEEE-1394 serial bus, then, are intended mostly for unidirectional digital data transmission (and bidirectional transmission of commands) into the computer for short distances, ranging from one-half to several meters. Typical functions of USB in the computer setup are presented in Figure 2.4. Figure 2.4 Typical functions of USB in the computer setup. Accordingly, the USB can be used for measuring data transmission; there is, however, no software for the measurement system with USB provided in standard resources of computer operating systems. The proprieties of the IEEE-1394 bus are more multipurpose, and, therefore, it is better fitted to the measurement system construction. The junctions of the USB and the IEEE-1394 buses are used in measurement systems to connect the following instruments: • An interface board, which, along with the computer, sets up the interface controller. The board of GPIB-USB or GPIB-1394 type (manufactured by National Instruments) may be an example of such a board. It can fulfill the controller functions of the IEEE-488 system.
- 47. Computer-Based Measurement Systems 35 • A DAQ board, which contains an analog-to-digital converter, and often a digital-to-analog converter. The DAQ board with a PC and a program creates a virtual measurement instrument. When all junctions of the main bus are installed inside the PC casing, the socket of the USB or of the IEEE-1394 is able to connect other devices to the computer, including the DAQ board. The USB is present in the structure of every PC or laptop. The desktop computer is equipped with two or more USB junctions, placed onto its casing, and the laptop is equipped with one or two junctions. The USB is designed to connect a variety of peripheral devices to the computer. The USB assures a standardiza- tion of cabling connecting these devices with the computer and serves for com- munication with them, giving a low number of interrups and input/output addresses. In its equipment part, the USB consists of host controller/root hub, USB concentrators, and USB devices, as shown in Figure 2.4. Both peripheral devices and hubs can be connected to the USB junction in the computer. The sending or receiving of data by a concrete USB device cannot take place on the initiative of this device, but only in consequence of periodic polling of all devices by the main controller. The USB concentrator is also a distributor and a signal amplifier. For example, hub 4 in Figure 2.5 fulfills only functions of the amplifier. Figure 2.5 The tree structure of USB devices in the computer system. The following four kinds of data transfer are possible in the USB: • Control transfers, applied after connecting a new USB device to the USB bus in order to configure the device. • Interrupts, related to periodical polling of slow “devices” (e.g., keyboards).
- 48. Measurement Systems and Sensors 36 • Bulk data transfers, applied in the case of devices with irregular communication, but with no transfer priority and no guarantee of bandwidth. Bulk transfers are designed to transfer large amounts of data with error-free delivery. The USB host will schedule bulk transfers after the other transfer types have been required. • Isochronous transfers, referring to devices working in real time (e.g., CD recorders and readers). A system transmits data in the asynchronous mode; however, breaks between particular signs must be integral to the multiplicity of the bit cycle Tb. It requires good synchronization of the transmitter and of the receiver; in the asynchronous transmission, character spacings are arbitrarily long. The features of the USB are as follows: • One type of interruptions and one USB address space. • The possibility of joining up to 127 devices. • One type of junction and cable for USB devices. The USB cable is a four- wire cable: two signal lines (for the differential transmission of a single signal) and two power supply wires. The maximum cable length amounts to 5m. • Low transmission rate less than 1.5 Mbps (by the USB 1.0 standard), the average rate less than 12 Mbps (by USB 1.1), high rate less than 480 Mbps (by USB 2.0), very high rate less than 5 Gbps (by USB 3.0) and very high rate plus – to 10 Gbps (by USB 3.1). • The installation in plug-and-play mode. • The possibility of providing power to peripheral USB devices from the computer through the bus. The USB port in the computer contains the power supply voltage of 5V, with a load-carrying capacity of 0.5A for external devices. A comfortable plug-and-play feature results in computer insensibility for joining and disconnecting USB devices on the bus. After such change of the system configuration the computer need not be restarted. The start of servicing (the initialization) of the USB bus by the computer follows automatically. Under the initialization of the computer or the main controller, the USB system (root hub) gives a 7-bit-wise address to a USB device and receives data concerning, among other elements, transmission mode and rates. Binary signals in the USB bus are transferred with a pair of wires appointed by D+ and D. The potential difference between the bus wires means: • Logical 1 for V(D+) V(D) > 200 mV. • Logical 0 for V(D+) V(D) < 200 mV.
- 49. Computer-Based Measurement Systems 37 In steady logical state, the voltage on the D+ or D line must be higher than 0.8V. Data Transfer by USB Transfer of data and commands by USB is carried out serially, bit by bit, in communication frames called packets. There are four types of USB packets: • Data packet. It begins with synchronization bits and a packet identifier (PID). Then follow the address, the endpoint field and, of course, the data field (to 1023 bytes)—see Figure 2.6 and the description below. Correctness of a transfer of the data packet is checked with a 16-bit CRC polynomial. • Token packet. It is sent by a host and used to query the device. It is formed by a PID, the address, the endpoint, and a 5-bit CRC polynomial. • Start of frame packet (SOF). The SOF packet is sent by the host as the flag of a control frame. The frames serve for checking of requests for bandwidth and for types of transfer in USB pipes (pipe–physical path between the host and the USB device). The SOF packet consists of a packet identifier, an 11- bit counter (it is incremented one per frame and indicates its number) and a 5-bit CRC polynomial. • Handshake packet. There are three types of handshake packets: ACKnowledgment, NAK (the function is not ready for data exchange), and STALL (the function is busy or an error occurred). The handshake packet includes only one field—the packet identifier. USB packets include the fields of bits listed below and shown in Figure 2.6: • Synchronization field, Sync, 8 bits. Synchronization bits are used to fit (to correct) their timing. • Packet identifier, PID, 8 bits. This field is divided into two groups of bits. Four bits determine the type and the format of the data transferred in the packet. The remaining bits indicate the type of the packet: data, token, SOF or handshake. • Address. This indicates the used pipe. • Endpoint, 4 bits; it indicates the endpoint either of data source or data receiver. • Field of data. This includes from 0 to 1023 bytes of data. Figure 2.6 A format of a packet for data transfer by USB. During the transmission of coded signs or numbers, bits are transferred beginning from the least significant bit (LSB).
- 50. Measurement Systems and Sensors 38 USB Junctions Four types of USB junctions are shown in Figure 2.7. The dimensions of USB junctions (plugs) are presented in Table 2.3. The line description in USB junctions is given in Table 2.4. Figure 2.7 Four types of USB junctions: (a) USB plugs (not to scale, from left): type A, type B, miniplug and microplug, and (b) plugs: A, B, and miniature (mini). Table 2.3 Dimensions of USB Junctions Type of junction A-plug B-plug Miniplug Microplug Width 12 mm 8.45 mm 7 mm 7 mm Hight 4.5 mm 7.78 3 mm 1.8 mm Table 2.4 Line Description in USB 1.0 and 2.0 Junctions Pin Number (A and B Type) Line Description Pin Number in Miniature Junction, B type 1 Power, +V 1 2 D, signal line 2 3 D+, signal line 3 4 GND 5 − ID (typically: not connected) 4 The standard USB 1.1 is not in use as an interface bus of the measurement systems due to slightly higher transmission rates (12 Mbps) as compared with the
- 51. Computer-Based Measurement Systems 39 system of the RS-485 serial interface (10 Mbps). Other signal parameters of the USB 1.1 are even worse than signals in the RS-485 system. Measurement Systems with USB The USB junction can serve for connecting the interface board to the computer (e.g., the IEEE-488 interface board), as shown in Figure 2.8. Figure 2.8 Connecting of the measurement system with the IEEE-488 parallel interface to the computer using the USB serial bus. As a result of connecting such a board, the computer becomes an interface controller of the IEEE-488. Particular devices attached to the USB system can exchange data in pairs, with different transmission rates for every pair. High speed of data transmission through the USB serial bus can be obtained only when both devices communicating with one another are prepared for such a transmission rate. The highest transmission rate in USB, amounting to 480 Mbps, is presently used for data exchange between memory disks or for reading DVD disk recordings. The USB 2.0 can be used as a measurement system bus with good dynamics, as shown in Figure 2.9. USB 2.0 ports have been installed in following instruments: • Oscilloscopes TDS 6000 and TDS 7000 series (Tektronix); • Oscilloscopes WaveSurfer series 400 and WaveRunner series 6000 (LeCroy); • Many types of spectrum analyzers manufactured by Rohde & Schwarz (e.g., R&S FS300); • Oscilloscopes Infiniium 54800, and arbitrary generators 33220A manufactured by Agilent Technologies. The computer-based measurement system with the USB bus can contain more instruments than the USB sockets installed in the computer because it is possible to include hubs in any point of the USB network. One can expect that the development of the USB equipment will be followed by the development of software for measurement systems with the USB. Most types of printers manufactured can be controlled from the computer by the USB bus installed in
- 52. Measurement Systems and Sensors 40 the printers (since the last 5 years the Centronics interface has been replaced by the USB port). Figure 2.9 The measurement system with the USB 2.0. The next examples of USB devices used for measurement systems are data acquisition cards offered by National Instruments. The NI USB 6009 device is a 14-bit analog-to-digital (A/D) converter with a sampling rate of 48 kS/s (S/s = samples per second), see Figure 2.10. Less advanced and cheaper is the NI USB 6008 device with a 12-bit A/D converter. It operates with a sampling rate of 10 kS/s. Both devices have eight analog inputs for voltage signals to be converted. The NI USB 6008/6009 devices can also generate an analog voltage (with 150 S/s rate) under a 12-bit control signal. A data acquisition card, connected to a computer via USB, can create a virtual instrument (see Chapter 13). Figure 2.10 The data acquisition card equipped with USB interface . 2.5 IEEE-1394 SERIAL BUS The serial bus with high transmission speed was developed in 1986 by Apple Computer under the name of FireWire; this name is still in use with the Apple company. In 1995, this bus was given the status of a standard, with the name
- 53. Computer-Based Measurement Systems 41 IEEE-1394 (other names of this bus are iLink or Digital Link). The IEEE-1394 bus fulfills a function similar to the function of the USB. It is intended, however, for devices requiring higher transmission rates, like digital cameras, DVD readers, measurement instruments, and navigational or medical instruments. The IEEE- 1394 and USB standards are complementary, but not interexchangeable. More recent computers are equipped with ports of both buses. Drivers servicing them are contained in the Windows 7, 8, and 10 operating system. The IEEE-1394 bus assures the highest transmission rate of all the serial interface standards. • 400 Mbps in a basic IEEE-1394a version. • 800 Mbps in the IEEE-1394b version . Work continues to improve the IEEE-1394b version in order to increase the transmission rate up to 3.2 Gbps. For such a high transmission rate, optical fiber lines will be required. The bus is intended for both modules installed in the PC casing and devices connected to the PC by means of a cable. Three types of IEEE-1394 junctions are used: nine-pin, six-pin, and four-pin junctions. Plug-in sockets of six-pin and four-pin junctions are shown in Figure 2.10. The descriptions of lines in the IEEE-1394 junctions are given in Table 2.5. 1 2 3 4 5 6 1 2 3 4 Figure 2.11 Two types of IEEE-1394 junctions (six-pin and four-pin plug-in socket). The features of the IEEE-1394 serial bus are as follows: • One type of interruption and one address space. • The possibility of joining up to 63 device. • High speed of transmission less than 800 Mbps. • The installation plug-and-play. • One (from three) chosen type of junction and cable for IEEE-1394 devices; the cable is usually a six-wire cable: four signal lines and two wires of power. A junction provides amounts of power on the IEEE-1394 bus up to 45W, with a maximum of 1.5A and 30V. Only nine-pin and six-pin junctions can carry power; four-pin junctions cannot. The general structure of the IEEE-1394 network (up to 64 devices) is a tree structure. The maximum length of a single section of the cable amounts to 4.5m. Up to 16 IEEE-1394 devices can be connected in chain mode, which gives the length of the bus equal to 67m for each chain, assuming the maximum length of each section of the cable.
- 54. Measurement Systems and Sensors 42 Table 2.5 Line Description in IEEE-1394 (FireWire) Junctions Junction A Color of a Wire ↔ Junction B Four-pin Six-pin Line description Line description Six-pin Four-pin - 1 Power, +V white Power, +V 1 - - 2 GND black GND 2 - 1 3 TPB red TPA 5 3 2 4 TPB+ green TPA+ 6 4 3 5 TPA orange TPB 3 1 4 6 TPA+ blue TPB+ 4 2 Figure 2.12 Three layers for IEEE-1394 protocol. The transmission protocol in an IEEE-1394 system is organized on three layers: a transaction layer, a link layer, and a physical layer, as shown in Figure 2.11. In the IEEE-1394 bus, two kinds of transmission are possible:
- 55. Another Random Scribd Document with Unrelated Content
- 56. “Is Mr. Atkins often away?” “Yes; he’s out of town every week or so, on business.” “Thank you, Mrs. Atkins, that is all,” the Coroner concluded, politely. But the lady was not so easily appeased, and flounced out of the room without deigning to glance at any of us. The detective slipped out after her—to call the maids, as he explained, but it was five or six minutes before he returned with the waitress. After answering several unimportant questions, the girl was asked whether she had ever seen the deceased before. “No, sir,” she replied, promptly. “Did anyone call on your mistress on Tuesday evening?” “I can’t say, sir; I was out.” “At what time did you go out?” “At about a quarter to eight, sir.” “Where did you go to?” “We went to a party at me sister’s.” “Who do you mean by ‘we’?” “The cook and me, sir.” “Ah, the cook went out, too?” “Yes, sir.” “Do you usually go out together?” “No, sir.” “How did it happen that you did so on Tuesday?”
- 57. “Mr. Atkins, he was away, so Mrs. Atkins she said we might both go out.” “Mr. Atkins is often away from home, isn’t he?” “Yes, sir.” “How often?” “About once a fortnight, sir.” “Has Mrs. Atkins ever allowed you both to go out together before?” “No, sir.” “Where does your sister live, and what is her name?” “Mrs. Moriarty, 300 Third Avenue.” The Coroner paused to scribble down the address, then resumed: “At what time did you get back from the party?” The girl tugged at her dress in some embarrassment. “It might have been after eleven,” she reluctantly admitted. “How much after—quarter past, half-past?” he suggested, as she still hesitated. “It was almost half-past, sir.” “And when you returned, did you see your mistress?” “Oh, yes, sir.” “Was she alone?” “Yes, sir,” the girl answered, with some surprise. “Did you notice anything unusual about her?” “Well, sir, she’d been crying, and I never see her cry before.”
- 58. “What did Mrs. Atkins say to you?” “She scolded us for being so late,” the girl answered shamefacedly. “Was that all she said?” “Yes, sir.” “Where was your mistress when you saw her?” “She was lying on the sofy in her bed-room, tired like.” “What did Mrs. Atkins do yesterday?” “She went out after breakfast and didn’t come back till nearly six.” “How did she seem when she returned?” “She’d been crying awful, and she just lay quiet and wouldn’t eat no dinner.” “Do Mr. and Mrs. Atkins get along well together?” “Oh, sir, they’re that loving,” she answered with a blush and a smile. Again my curiosity got the better of my discretion, and I asked: “Did you hear any strange noises during the night?” The Coroner glared at me, but said nothing this time. “Well,” replied the girl, “me and Jane did think as we’d heard a scream.” Ha, ha, thought I, and I saw Mr. Merritt indulge in one of his quiet smiles. “So you heard a scream,” said the Coroner. “I don’t know for sure; I thought so.”
- 59. “At what time did you hear it?” “I don’t know, sir; some time in the night.” “What did you do when you heard it?” “Nothing, sir.” This was all that could be got out of her, so she made way for the cook, who, after being cross-questioned at some length, did no more than corroborate the waitress’s statement, only she was more positive of having heard the “screech” as she called it. “Could you tell whether it was a man or woman who screamed?” inquired the Coroner. “It was a woman’s voice, sir.” Mr. Stuart, who was next admitted, proved to be a small, middle- aged man, extremely well groomed, and whom I recognized as one of the members of my Club, whose name I had never known. On being asked if he had ever seen the dead man before, he solemnly inserted a single eye-glass into his right eye, and contemplated the corpse with the greatest imperturbability. “So far as I can remember, I have never seen the man before,” he answered at last. After replying satisfactorily to a few more questions, he was allowed to retire, and his cook took his place. She was a large, stout woman about thirty years old, with a good deal of that coarse Southern beauty, which consists chiefly in snapping black eyes, masses of dark hair, and good teeth. On catching sight of the corpse, she threw up her hands and uttered a succession of squeals, which she seemed to consider due to the horror of the occasion, and then turned serenely towards the Coroner, and with a slight courtesy stood smilingly awaiting his questions. “What is your name?” he inquired. “Jeanne Alexandrine Argot,” she replied.
- 60. “You are in the employ of Mr. Stuart?” “Yes, sar. I ’ave been with Mr. Stuah, six a years, and he tell you ——” “Please look at the deceased, and tell me if you have ever seen him before?” the Coroner hastily interrupted. “No, sar.” After answering a few more questions with overpowering volubility, she withdrew, and her husband entered. He was a tall, vigorous man, with large hawk-like eyes, apparently a good deal older than his wife. He bowed to us all on entering, and stood respectfully near the door, waiting to be spoken to. “What is your name?” inquired the Coroner. “Celestin Marie Argot.” “You work for Mr. Stuart?” “Yes, sar; I am Meester Stuah’s butlair.” “Look at this corpse, and tell me if you can identify it as that of any one you know, or have ever seen?” He now glanced for the first time at the body, and I thought I saw his face contract slightly. But the expression was so fleeting that I could not be sure of it, and when he raised his head a few moments later he seemed perfectly composed and answered calmly: “I do not know ze man.” Apparently the Coroner was not completely satisfied, for he went on: “You know that this man has been murdered, and that it is your duty to give us any information that might lead to his identification. Have you seen any suspicious persons about the building during the last few days?”
- 61. “No, sar; nobody,”—but I thought he had hesitated an instant before answering. “You must see a good many people pass up and down the back stairs,” the detective remarked; “especially in this hot weather, when you must be obliged to leave the kitchen door open a good deal so as to get a draught.” The man cast a hurried, and I thought an apprehensive, glance at Mr. Merritt, and replied quickly: “Yes, sar; ze door is open almos’ all ze time, but I ’ave seen nobody.” “Nobody?” repeated the detective. “Yes, sar,” Argot asserted, still more emphatically. “No vone, excep’ ze butchair, ze bakair, and ze ozer tradesmen, of course.” “How early are you likely to open the kitchen door? To leave it open, I mean?” “Oh, not till eight o’clock, perhap—Madame Argot, she stay in déshabille till zen.” “What time do you go to bed?” “At ten o’clock generally, but some time eleven o’clock—even midnight—it depens.” “What time did you go to bed on Tuesday?” “At eleven, sar.” “What had you been doing during the evening?” “I had been at a restaurant wiz some friends.” “And when did you return?” “At about half-pas’ ten.” “Did you come in the back way?”
- 62. “Yes, sar.” “How did you get in?” “My wife, she open ze door.” “And you saw nobody as you came in?” He paused almost imperceptibly. “No, sar,” he answered. But I was now convinced that he was holding something back. “Very well; you can go,” said the Coroner. The fellow bowed himself out with a good deal of quiet dignity. “I kinder fancy that man knows something he won’t tell,” said the Coroner. “Now, we’ve seen every one but the workmen,” he continued, wearily, mopping his forehead. “I don’t believe one of them knows a thing; still, I’ve got to go through with it, I suppose,” and going to the door he beckoned them all in. There were five of them, including the foreman, and they appeared to be quiet, respectable young men. After looking at the dead man intently for some minutes, they all asserted that they had never laid eyes on him before. “Now have any of you noticed during the three days you have been working here anybody who might have taken the key, kept it for some hours, and returned it without your noticing it?” inquired the Coroner. “We’ve seen no strangers,” the foreman replied, cautiously. “Who have you seen?” The foreman was evidently prepared for this question. “Well, sir, we’ve seen altogether six people: Jim, and Joe, and Tony, Mr. McGorry, Miss Derwent, and the Frinchman,” he replied, checking them off on his fingers. “When did the Frenchman come up here?”
- 63. “Yistidy morning, sir; he said he come to see the decorations, and he come again about three; but he didn’t stay long. I warn’t a-going to have him hanging round here interfering!” “Did any of his actions at the time strike you as suspicious?” “No, sir,” acknowledged the foreman. “And Miss Derwent; when did you see her?” “I didn’t see her myself in the morning, but he”—with a nod towards one of the men,—“he saw her look in as she was waiting for the elevator, and in the afternoon she come right in.” “Did she say anything?” “Yes, sir; she said the paint and papers were mighty pretty.” “When you saw Miss Derwent,” said the Coroner, addressing the man whom the foreman had pointed out, “what was she doing?” “She was standing just inside the hall.” “Was her hand on the door knob?” “I didn’t notice, sir.” “Did the young lady say anything?” “When she saw me a-looking at her, she just said: ‘How pretty!’ and went away.” “Have any of you seen Mr. or Mrs. Atkins, or either of their girls, since you have been working here?” They all replied in the negative. The Coroner’s physician turned up at this juncture, with many apologies for his late arrival, so, having no further excuse for remaining, I took my leave. The lower hall swarmed with innumerable reporters, trying to force their way upstairs, and who were only prevented from doing so by the infuriated McGorry and two or three stalwart policemen. On catching sight of me they all fell
- 64. upon me with one accord, and I only managed to escape by giving them the most detailed description of the corpse and professing complete ignorance as to everything else.
- 65. W CHAPTER VI A LETTER AND ITS ANSWER HEN I got back to my diggings I was astonished to find that it was only ten o’clock. How little time it takes to change the whole world for one! All day long I forced myself to go about my usual work, but the thought of May Derwent never left me. It was the greatest relief to find that in none of the evening papers did her name appear. How McGorry managed to conceal from the reporters the fact that she had been in the building remains a mystery to this day—but how thankful I was that he was able to do so! Already my greatest preoccupation was to preserve her fair name from the least breath of scandal. Not for an instant did I believe her to be connected with the murder;—on the other hand, I felt equally sure that she was in some great trouble, the nature of which I could not even guess. I longed to protect and help her, but how was I to do so, ignorant as I was of everything concerning her. I didn’t even know where she was at that moment. At her mother’s, perhaps. But where was that? Suddenly I remembered that my great friend, Fred Cowper, had mentioned in one of his recent letters that Mrs. Derwent and his mother were near neighbours in the country. To think that that lucky dog had been spending the last month within a stone’s throw, perhaps, of her house—had seen her every day probably, and had been allowed these inestimable privileges simply because he had broken an old leg! And I, who would gladly have sacrificed both legs to have been in his place, was forced to remain in New York because—forsooth!—of an apoplectic old patient —who refused either to live or die! Well, as I couldn’t go to her, it was at any rate a comfort to be able to get news of her so easily—so seizing a pen, I hastily scratched off the following note:
- 66. New York, August 10, 1898. Dear Fred: You know me pretty well and know therefore that I’m not a prying sort of fellow —don’t you? So that when I ask you to tell me all you know about Miss May Derwent—I hope you will believe that I am animated by no idle curiosity. A doctor is often forced to carry more secrets than a family solicitor, and is as much in honor bound. Through no fault of my own, I have come into the possession of certain facts relating to Miss Derwent which lead me to believe that she is in great trouble. Furthermore, I am convinced that I could help her, were I not handicapped by my very slight personal acquaintance with her, but more than that by my entire ignorance regarding certain details of her life. I might as well acknowledge that I am interested in the young lady, and am anxious to serve her if I can. But if I am to do so, I must first find out a few particulars of her life, and these I hope you can give me. In the first place I want to know whether she has any young male relative who is tall, with good figure? I remember hearing that she is an only child, but has she no cousin with whom she is on terms of brotherly intimacy? Secondly, Is she engaged, or reported to be engaged, and if so, to whom? Thirdly, What are the names of her most favored suitors? Fourthly, What lady does she know intimately who has very dark hair, and is also slight and tall? I don’t need to tell you to treat this letter as absolutely confidential, nor to assure you again that only the deepest interest in Miss Derwent, and the conviction that she is in need of help, induce me to pry into her affairs. More than this I cannot tell you, so don’t ask me. Good-night, old chap! Hope your leg is getting on all right. Affectionately yours, Charles K. Fortescue. Hope Farm, Beverley, L. I., Friday, August 11. Dear Charley,—You may imagine how exciting I found your letter when I tell you that I have known May Derwent since she was a tiny tot, and that their country place is not half a mile from here. She is exactly my sister Alice’s age, and I have
- 67. never known her very well till she came out last winter, for eight years make a big barrier between children. I like and admire May extremely, for not only is she a very beautiful girl, but an extremely nice one, as well. Difficult as it may be to explain certain things, I am sure that, whatever the trouble she is in, if you knew the whole truth, you would find it only redounded to her credit. She is an impulsive, warm-hearted and rather tempestuous child—generous, loyal, and truthful to a fault. I have just been discreetly sounding Alice about her, and asked why I had not seen May since I had been down here this time, as on former occasions she used always to be running in and out of the house. And Alice tells me that for the last three months May has been a changed being. From a happy, thoughtless girl, overflowing with health and spirits, she has become a listless, self-contained, almost morose woman. She refuses to go anywhere, and spends most of her time either in her own room or taking long solitary walks or rides. The doctor talks of nervous prostration, but do you think it likely that a vigorous, athletic young girl would develop nerves solely in consequence of a few months’ gaiety during the winter? It seems to me incredible, and so I am forced to believe that May has something on her mind which is reacting on her body, causing her to shun all the things she used to delight in. Now, when a young, rich, beautiful, and sought-after girl suddenly takes to avoiding her species, and becomes pale and melancholy, the usual explanation is—an unhappy love affair. And, of course, that may still turn out to be the truth in this case; but in the meantime I have another hypothesis to suggest, that seems to me to fit in with the known facts even better than the other. May Derwent is not an only child, but has, or at any rate had, a brother about ten years older than herself who, I confess, was one of the heroes of my childhood. Only a little older than the rest of us boys, he was much bigger and stronger. He was the leader of all our games, and the instigator of our most outrageous exploits. He was the horror of all parents and the delight of all children. Cruel, vindictive, untruthful, leaving others to pay the penalty for his faults whenever it was possible, he was not a nice boy even in those early days, but then he was so handsome, so bold and unscrupulous, so inspired in devising new crimes for us to commit, that it is hardly to be wondered at that he was at the same time our terror and our idol. His school record was bad; his college record was worse, till one fine day he suddenly and mysteriously disappeared from Harvard, and has never been heard of since. What had occurred I never could find out; that it was something very disgraceful I am sure, for his mother, whose pride and hope he had been, never again mentioned his name. Now, don’t you think it quite possible that he may have returned and been bothering his sister in some way? She may be either trying to shield him from still greater disgrace, or be endeavouring to spare her mother the further knowledge of his misdeeds. Mind you, these are all merely the wildest conjectures.
- 68. As for May’s lovers, their name is simply legion, including young Norman, the millionaire, Sir Arthur Trevor, Guy Weatherby and a painter chap—Greywood, I think his name is. Mère Derwent, I believe, favors Norman’s suit, having (sensible woman!) a great faith in American husbands, but there is a rumour that May, with the perversity of her sex, is inclined to smile on the young artist, who, I am told is an affected chap, just back from Paris, without either money or talent. But no doubt he strikes her as a more romantic lover than good old Norman, who is the best of fellows, and absolutely eligible in every way. Alice tells me that May has appeared quite eager for her Bar Harbor visit, notwithstanding that she has refused all other invitations, and Mrs. Derwent has had great hopes that the change would do her good. What you have told me is no small tax on my discretion, but what you have refrained from telling taxes my curiosity far more. But notice—I ask no questions!! By the way, why don’t you come down and spend next Sunday with us? You might see the lovely May again,—who knows? Affectionately yours, Fred.
- 69. F CHAPTER VII MR. MERRITT INSTRUCTS ME RED’S letter was a great relief to me. I had not dared to allow my thoughts to dwell on the man whom I had seen in May Derwent’s apartment on that eventful night. The supposition, however, that it was her brother, explained everything satisfactorily. Nothing could be more likely than that this angel of mercy should give shelter to this returned prodigal, and try to save him from the punishment he so richly deserved. But what cared I what he had done? She—she—was immaculate. At the hospital that morning, I was in such good spirits that I had some difficulty in keeping my elation within bounds. As it was, I noticed that several nurses eyed me with suspicion. My preoccupation about Miss Derwent’s affairs had been so great that I had hardly given a thought to the mysterious murder, and was consequently very much surprised, on returning home that afternoon, to find the detective patiently awaiting me. “Well, Mr. Merritt,” I exclaimed; “glad to see you; what can I do for you? Anything wrong with your heart, or your liver, or your nerves, eh?” “Well, Doctor, I guess my nerves are pretty near all right,” he answered, with a slow smile. “I’m glad to hear it. Won’t you sit down?” He selected a comfortable chair, and we sat down facing each other. I wondered what could be coming next.
- 70. “Now, Doctor,” he began, in a matter-of-fact voice, “I’d like you to tell me all you know of the murder.” He had taken me completely by surprise, but I am learning to control my features, and flatter myself that I did not move a muscle as I quietly replied: “This is a very strange question, and I can only answer that I know nothing.” “Oh, hardly as little as that,” the detective rejoined, with irritating complacency. “Just as little as that,” I asserted, with some warmth. “Well, Doctor, if that is the case, you can no doubt explain a few things that have been puzzling me. In the first place, will you tell me why, if you were not expecting another victim, you showed such surprise at the sight of the corpse? What reason could you have had for being so deeply interested in the relative positions of your roof— not your office, mind you, but your roof—and the room in which the body was found, unless you had noticed something unusual from that point of observation? Why were you so sure that the Derwent’s flat was occupied, if you had not seen some person or persons there? By the way, I noticed that from your roof I could look directly into their windows. Again, you betrayed great surprise when Miss Derwent lifted her veil. Why did you do so, except that you had previously seen a very different looking person in her apartment? And why did you select the Atkins’s two servants out of all the people in the building, to question about a certain noise, but that you yourself had heard a scream coming from their premises? And, lastly, you showed an unexplained interest in the back door of the Rosemere, which is particularly suggestive in view of the fact that this window is exactly opposite to it. I need only add that your presence on the roof during some part of Wednesday night, or early Thursday morning, is attested by the fact that I found some pipe- ash near the chimney. You smoke a pipe, I see” (pointing to a rack
- 71. full of them); “your janitor does not, neither do your two fellow- lodgers. Besides that, all the other occupants of this house are willing to swear that they have not been on the roof recently, and those ashes could not have been long where I found them; the wind would have scattered them. You see, I know very little, but I know enough to be sure that you know more.” I was perfectly dumbfounded, and gazed at the detective for some moments without speaking. “Well, granted that I was on the roof during a part of Wednesday night, what of it? And if I did hear or see anything suspicious, how can you prove it, and above all, how can you make me tell you of it?” “I can’t,” rejoined Mr. Merritt, cheerfully. “I can only ask you to do so.” “And if I refuse?” “Then I shall have to delay satisfying my curiosity till we meet in court, but I do not doubt that my patience will then be adequately rewarded, for a skilful lawyer will surely be able to get at many details that would escape me, and I hardly think that you would resort to perjury to shield two women whom I am convinced you never laid eyes on before yesterday, and have certainly not seen since.” The detective paused. I still hesitated, for I felt an extreme reluctance to further compromise that poor girl by anything I might say. “Come, Doctor,” he urged, leaning forward and placing his hand on my knee, “don’t you think it would be better for all parties for you to tell me what you know? I am as anxious to shield the innocent as you can be. By withholding valuable information you may force me to put a young lady through a very trying and public ordeal, which I am sure might be easily spared her, if I only knew a few more facts of the case.”
- 72. This last argument decided me, and making a virtue of necessity I gave him a minute account of all I had seen and heard. When I came to describing the man’s prolonged search Mr. Merritt nodded several times with great satisfaction. “Can’t you tell me a little more how this man looked?” he eagerly inquired. “You must have seen him pretty clearly while he was moving around that lighted room. Had he any hair on his face?” “Well,” I confessed, “it is a funny thing, but I can’t for the life of me remember; I’ve tried to; sometimes I think he was clean shaven, and again I am sure he had a small moustache.” The detective glared at me for a moment; it was difficult for him to forgive such aggravating lack of memory. To be given such an opportunity and to foozel it! He heaved a sigh of resignation as he inquired: “Can you remember how he was dressed?” “Oh, yes,” I replied with alacrity, anxious to retrieve myself, “he had on a white shirt and dark trousers, and his sleeves were rolled back.” “Did he close the windows before he left?” “Yes, and he pulled down the blinds also.” “You are sure that you saw no one in the apartment resembling Miss Derwent?” “Quite sure; the woman I saw was taller and had flat, black hair.” “What do you mean by ‘flat’?” “Why, nowadays girls wear their hair loose; it bulges away from their faces; but hers lay tight to her head in a flat, black mass,” I explained.
- 73. I then harped on the probability of the return of Miss May’s prodigal brother, and suggested the possibility that the dark-haired woman might be his wife. “Well, well, Doctor! This is all very interesting. The story of the brother, especially. You see, I had already discovered that a man had spent many hours in her apartment——” “How did you find that out?” I interrupted. “Oh, quite easily,” rejoined the detective; “as soon as all the excitement was over yesterday, I made McGorry open the Derwent’s apartments for me. You may imagine what a fuss he made about it. Well anyhow he got me——” “But why did you want to get in?” I inquired; “did you suspect her?” “No,” he replied, “I did not. But in my profession you take no chances. Impressions, intuitions, are often of great value, only you must be careful always to verify them. I was almost sure that the young lady was innocent, but it was my business to prove her so. Now, it is certain that the person, or persons, who smuggled the corpse into the room where it was found, must, at one time or another, have had the key of that apartment in their possession, and there are only three people whom we know of as yet who were in a position to have had it. These three are: Miss Derwent, the French butler, and, of course, McGorry. So far I have not been able to connect the latter two, even in the most indirect way, with the catastrophe. Unfortunately, that is not the case with the young lady. One person, at least, has identified the body as that of her visitor, and your behaviour,” he added, with a smile, “led me to believe that you suspected her of something. Not of the crime, I felt sure of that, but of what, then? I determined to find out, and now that I have done so, let me tell you that I am still convinced of her innocence.” I jumped up and shook him by the hand. “So am I, so am I,” I exclaimed.
- 74. “But this is a very queer case,” he continued, “and I shall need all the assistance you can give me, if——” “You shall have it,” I broke in, enthusiastically; “anything I can do. But tell me, first, how you found out about Miss Derwent’s brother?” “Not so fast, young man! At present, we know nothing about a brother. I only said that I had discovered in the apartment traces of the recent and prolonged presence of a man, and I may add of a man of some means.” “How did you find that out? Especially about his means?” I inquired, with a smile. “Quite easily. In the parlor, which was the first room I entered, I noticed that every piece of furniture had been lately moved from its place. Now, this was too heavy a job for a girl to have undertaken single-handed. Who helped her, I wondered? Her visitor of Tuesday evening might have been the person, but for various reasons I was inclined to doubt it. I thought it more likely to have been the woman whose existence your behaviour had led me to infer. I next examined the dining-room. A few crumbs showed that it had been used, but I could find no traces of her mysterious companion. The library had not even been entered. On the floor above, the front bedroom alone showed signs of recent occupation. Two crumpled sheets were still on the bed, and in the drawers were several articles of woman’s apparel. Returning to the lower floor by the back stairs, I found myself in the kitchen. Here, in the most unexpected place, I discovered an important clue.” Mr. Merritt paused, and looked at me with a gleam of triumph in his eye. “Yes, yes, and what was that?” I inquired, breathlessly. “Only the odor, the very faintest ghost of an odor, I may say, of cigar-smoke.” “In the kitchen?” I exclaimed, incredulously.
- 75. “In the kitchen,” repeated the detective. “I at once drew up the blinds, and looked out. The window opened directly on the fire escape, with nothing opposite but the roofs of some low houses. Pulling out my magnifying glass, I crawled out. I soon satisfied myself that the stairs leading up and down had not been recently used; on the other hand, I was equally sure that someone had very lately been out on the small landing. So I sat down there and looked about me. I could see nothing. At last, by peering through the bars of the iron flooring, I thought I could discern a small brown object, caught in between the slats of the landing below. I climbed down there mighty quick, I can tell you, and in a moment held the butt end of a cigar in my hand. It was, as I had suspected, from the delicate odor it had left behind, one which had cost about fifty cents. I now extended my search downward, and examined every window- sill, every crevice, till I reached the basement, and, as a result of my hunt, I collected five cigar stumps, all of the same brand. From the number, I concluded that whoever had been in the apartment had been there a considerable time. From his only smoking in the kitchen or on the fire-escape, I gathered that he was anxious to leave no traces of his presence; and lastly, from the quality of his cigars, I judged him to be a man of means. So you see I had discovered, even without your assistance, that, although Miss Derwent may have told us the truth, she certainly had not told us all of it.” I nodded gloomily. “What you tell me of this dark-haired woman is still more puzzling,” the detective continued. “She has covered up her tracks so well that not only did I find no trace of her, but no one, not even yourself, saw her either enter or leave the building. And I should never have dreamed of her existence if I had not noticed your surprise when Miss Derwent lifted her veil. Now, the first thing to be done is to try and find this strange couple, and we will begin by tracing the man whom you saw leaving the Rosemere with a market- basket. It will be easy enough to find out if he is nothing but a local
- 76. tradesman, and if he is not, then in all probability he is the man we want. The detective who is watching Miss Derwent——” “A detective watching Miss Derwent!” I exclaimed. “Why, yes. What did you expect? I sent one down with her to the country yesterday.” Perhaps I ought to have been prepared for it, but the idea of a common fellow dogging May Derwent’s footsteps, was quite a shock to me, so I inquired, with considerable ill-humor: “And what does he report?” “Nothing much. The young lady returned to her mother, as she said she would, and since then has kept to her room, but has refused to see a doctor.” “Have you discovered yet who the dead man really is?” I asked, after a slight pause. “No,” answered the detective, with a troubled look, “and I can’t make it out. Jim and Joe each persists in his own identification. I expected Jim to weaken, he seemed so much less positive at first, but whether he has talked himself into the belief that the corpse is that of the young lady’s visitor, or whether it really does resemble him so much as to give the boy grounds for thinking so, I can’t make out.” “I see, however, that you believe the murdered man to be Mrs. Atkins’s friend, of whose history and whereabouts she was so strangely ignorant.” “Well, I don’t know,” the detective replied. “We have found out that an Allan Brown did engage a berth on the midnight train to Boston.” “Really? Why, I was sure that Allan Brown was a creation of the little lady’s imagination. By the way, it is a strange coincidence that two mysterious Allans are connected with this case.”
- 77. “Yes, I have thought of that,” the detective murmured; “and Allan is no common name, either. But it is a still stranger circumstance that neither of Allan Brown nor of the murdered man (I am now taking for granted that they are not identical) can we discover the slightest trace beyond the solitary fact that an upper berth on the Boston train was bought on Tuesday afternoon, by a person giving the former’s name, and whose description applies, of course, equally to both. Mrs. Atkins volunteers the information that Brown was a stranger in the city, and so far I have no reason to doubt it. Now, a man who can afford to wear a dress suit, and who is a friend of a woman like Mrs. Atkins, presumably had fairly decent quarters while he was in town. And yet inquiries have been made at every hotel and boarding-house, from the cheapest to the most expensive, and not one of them knows anything of an Allan Brown, nor do they recognize his description as applying to any of their late guests. The deceased, of course, may have had rooms somewhere, or a flat, or even a house, in which case it will take longer to trace him; although even so, it is remarkable that after such wide publicity has been given to his description, no one has come forward and reported him as missing. The morgue has been crowded with idle sightseers, but nobody as yet claims to have seen the victim before.” “That is queer,” I assented, “especially as the dead man was in all probability a person of some prominence. He certainly must have been rich. The pearl studs he wore were very fine.” “Oh, those were imitation pearls,” said the detective, “and I am inclined to think that, far from being wealthy, he was, at the time of his death, extremely badly off, although other indications point to his having seen better days.” “Really!” I exclaimed. “Yes; didn’t you notice that his clothes, although evidently expensive, were all decidedly shabby? That his silk socks were almost worn out; that his pumps were down at the heel?”
- 78. “Yes, I did notice something of the kind.” “But those large imitation pearls blinded you to everything else, I see,” Mr. Merritt remarked, with a smile. “I suppose so,” I acknowledged; “they and the sleeve-links with the crest.” “Ah, those are really interesting, and for the first time in my life I find myself wishing that we were more careful in this country about the use of such things. Unfortunately, we are so promiscuous and casual in adopting any coat-of-arms that happens to strike our fancy that the links become almost valueless as a clue. Still, I have sent one of them to an authority in heraldry, and shall be much interested to hear what he has to say about it. By the way, did anything else strike you as peculiar about the corpse?” “No,” I answered, after a moment’s reflection. “It did not seem to you odd that no hat was found with the body?” “Dear me! I never noticed that. How singular! What could have become of it?” “Ah, if we only knew that we should be in a fair way to solving this mystery. For I have found out that, whereas the description of Miss Derwent’s visitor and Mrs. Atkins’s friend tally on all other points, they differ radically on this one. The former wore a panama, whereas the latter wore an ordinary straw hat. Now, one of those hats must be somewhere in the Rosemere, and yet I can’t find it.” “Mr. Merritt,” I inquired, “have you any theory as to the motive of this murder?” “Not as yet,” he replied. “It may have been jealousy, revenge, or a desire to be rid of a dangerous enemy, and if you had not given it as your opinion that the man met his death while wholly or semi- unconscious, I should have added self-defence to my list of
- 79. possibilities. The only thing I am pretty sure of is—that the motive was not robbery.” “Look here, Mr. Merritt, I can’t help wondering that, whereas you have treated Miss Derwent with the utmost suspicion, have made a thorough search of her apartment, and have even sent a sleuth to watch her, yet you have shown such indifference to Mrs. Atkins’s movements. Surely suspicion points quite as strongly to her as to the young lady?” “No, it doesn’t,” replied the detective. “The key! You forget the key cannot so far be connected with her. But, may I ask, who told you that I had neglected to make inquiries about the lady?” “Nobody; I only inferred,” I stammered. “You were wrong,” continued Mr. Merritt. “I have made every possible inquiry about Mrs. Atkins. I have even sent a man to Chicago to find out further particulars, although I have already collected a good deal of interesting information about the little lady’s past life.” “Really? And was there anything peculiar about it?” “No; I can’t exactly say there was. Mrs. Atkins is the only daughter of a wealthy saloon-keeper, John Day by name, and is twenty-six years old. Nothing is known against her except that in that city she chose her companions from amongst a very fast crowd. There is also a rumor, which the Chicago detective has not been able to verify, that when she was about sixteen or seventeen years old, she eloped with an Eastern man, from whom she was almost immediately divorced. At any rate, she has been known for a good many years as Miss Day, and has lived at home with her father. The memory of her marriage, if indeed she ever was married, has grown so dim that a great many people, among whom may be numbered some of her intimate friends, have never heard of it, and vehemently deny the whole story. I hope, however, soon to find out the facts of the case. Young Atkins met his wife last winter at Atlantic City, and at once fell
- 80. in love with her. His father, who is a very wealthy contractor, was strongly opposed to the match. He was very ambitious for his son, and thought the daughter of a saloon-keeper, whose reputation was none of the best, was no desirable wife for his boy.” “But they married in spite of him,” I said. “Yes, and old man Atkins has become reconciled to them, and makes them a very handsome allowance.” “How long have they been married?” I asked. “Since the fifteenth of April,” replied the detective, “and they were not married in Chicago, but in this city. I guess the lady was not over anxious to introduce her husband to her former pals.” “I suppose you have searched her apartment for a possible clue,— the hat, for instance?” “Yes, but as she has not been out since Wednesday, I have not been able to make as thorough a search as I should like. She is a shy bird, and I don’t want to frighten her till I have a few more facts to go on. If she thinks herself watched she may become wary, while now, I hope she will make use of her fancied security to do something which may give us a lead.” “Well, Mr. Merritt, I conclude from all this that, although you are unable to trace the possession of the key to Mrs. Atkins, nevertheless, your suspicions point towards her?” “Certainly not. There is nothing to connect her with the tragedy, except the fact that one negro boy identified the corpse as that of one of her visitors. On the contrary, the more I look into this case, the less do I see how the lady could be involved in it. Let us suppose that she did kill the man. Where could she have secreted him during the twenty-four hours that must have elapsed before the body was finally disposed of? The only place of concealment on the lower floor of her apartment is a coat closet under the stairs, and I doubt very
- 81. much whether a small, unmuscular woman like Mrs. Atkins is capable of dragging so large a man even for a short distance.” “But,” I suggested, “the murder may have been committed in the hall, just a step from this hiding-place.” “Yes, that is, of course, possible. But there is still another objection. The closet is so small that I do not believe a man could be got into it without doubling him up, and of that the body shows no signs. Besides, if Mrs. Atkins is guilty, we must believe her husband to be her accomplice, for who else could have helped her hide her victim? Now, you must know that the Atkins men, both father and son, bear most excellent reputations, especially the young man, of whom every one speaks in the highest terms, and I do not think that a person unaccustomed to deceit could have behaved with such perfect composure in the presence of a corpse of which he had criminal knowledge.” “But he did show some emotion,” I urged. “Oh, yes; I know what you mean,—when he learned that the man was murdered on Tuesday night he seemed startled.” “Well, how do you account for that?” “I don’t account for it. Why, Doctor, in a case like this there are a hundred things I can’t account for. For instance, what was the cause of Mrs. Atkins’s scream? You have no idea; neither have I. Why did she show such emotion at the sight of the corpse? I am not prepared to say. Why did she appear so relieved when she heard that the murder occurred on Tuesday? I can formulate no plausible explanation for it. And these are only a few of the rocks that I am running up against all the time.” “But look here. If you really believe Miss Derwent and Mrs. Atkins both innocent, who do you think killed the man?” “I don’t know. Oh, I am aware that the detective of fiction is always supposed to be omniscient, but my profession, Doctor, is just
- 82. like any other. There is no hocus-pocus about it. To succeed in it requires, in the first place, accurate and most minute powers of observation, unlimited patience, the capacity for putting two and two together. Add to this an unprejudiced mind, and last, but not least, respect, amounting to reverence, for any established fact. Now, the only facts we have as yet gathered about this murder are: that the man was young, dissipated, and was stabbed through the heart by some very small instrument or weapon; that his assailant was an inmate of the Rosemere; that the crime was committed on Tuesday night; and, lastly, that whoever placed the body where it was found must, at one time or another, have had the key to the outside door in his or her possession. Whatever else we may think or believe, is purely speculative. We presume, for instance, that the man was poor. As for the other facts we have gleaned about the different inmates of the building, till we know which one of them had a hand in this tragedy, we cannot consider what we have learned about them as throwing any light on the murder. About that, as I said before, we know mighty little, and even that little is the result of thirty-eight hours’ work, not of one man alone, but of seven or eight.” “Indeed!” I exclaimed. “Now, both ladies deny that they knew the deceased, and perhaps they are right. It is, of course, possible that there was a third man in the building that evening, who was also tall, dark, and wore a pointed beard. It is not likely, however. Such a coincidence is almost unheard of. Still it is possible, and that possibility must be reckoned with. Now, I must be off,” said Mr. Merritt, rising abruptly from his chair, “and if you hear any more of the young lady’s movements, let me know. There’s my address. In the meantime, thank you very much for what you have already told me.” And before I could get out one of the twenty questions that were still burning on my lips, the man was gone. For some minutes I sat quite still, too miserable to think connectedly. Alas! my fears had not been groundless. The poor girl
- 83. was in even greater trouble than I had supposed. I believed the detective to be a decent chap, who would keep his mouth shut, but how dreadful to think that her reputation depended on the discretion of any man. Should it become known that she had received one young man alone in an empty apartment, while another was seen there at three o’clock in the morning, it would mean social death to her. Oh, for the right to offer her my protection, my services! Of course, it was now absolutely necessary to trace the man who spent Tuesday evening with her, and to prove beyond doubt that he was still alive. I wished that this might be done without her knowledge, so as to spare her the shock of finding herself suspected of a crime. Again I thought of Fred, and at once sent him a few lines, begging him to let me know whether he or his sister knew of any friend or admirer of Miss Derwent who resembled the enclosed description, and if either of them did know of such a person, please to telegraph me the man’s name, and, if possible, his address. While giving no reasons for my questions, I again enjoined the greatest secrecy.
- 85. CHAPTER VIII AN IDENTIFICATION Telegram. Dr. Charles Fortescue, Madison Avenue, New York City. Saturday, August 12. Maurice Greywood. Can’t find his address. May be in Directory. Frederic Cowper. Clipping from the New York Bugle, Sunday, August 13. Landlady Identifies Body of the Rosemere Victim as that of her vanished lodger, artist Greywood. Police still Sceptical. Mr. Maurice Greywood, the talented young artist who returned from Paris the beginning of last winter, has disappeared, and grave fears for his safety are entertained. He was last seen in his studio, 188 Washington Square, early on Tuesday, August 8th, by Mrs. Kate Mulroy, the janitress. Ever since the young artist moved into the building, Mrs. Mulroy has taken complete charge of his rooms, but, owing to a disagreement which took place between them last Tuesday, she has ceased these attentions. Yesterday evening, while looking over a copy of the Bugle of the preceding day, Mrs. Mulroy came across the portrait of the unknown man whose murdered body was discovered under very mysterious circumstances in an unoccupied apartment of the Rosemere, corner of —— Street and Madison Avenue, on the preceding Thursday. She at once recognized it as bearing a striking resemblance to her lodger. Thoroughly alarmed she decided to investigate the matter. After knocking several times at Mr. Greywood’s door, without receiving an answer, she opened it by means of a pass-key. Both the studio and bedroom were in the greatest confusion, and from the amount of dust that had accumulated over everything, she concluded that the premises had not been entered for several days. Her worst fears being thus confirmed, she hastened at once to the Morgue,
- 86. and requested to see the body of the Rosemere victim, which she immediately identified as that of Maurice Greywood. Strangely enough, the police throw doubts on this identification, although they acknowledge that they have no other clue to go on. However, Mrs. Greywood, the young man’s mother, has been sent for, and is expected to arrive to-morrow from Maine, where she is spending the summer. The people at the Rosemere are still foolishly trying to make a mystery of the murder, and refuse all information [etc., etc.]. To Dr. Charles K. Fortescue from Dr. Frederic Cowper, Beverley, L. I. Sunday Evening, August 13th. Dear Charley: No sooner had I read in to-day’s paper that the body found in the Rosemere had been identified as that of Maurice Greywood, than I knew at once why you have taken such an interest in poor May. I see now that you have suspected from the first that the murdered man was not unknown to her, and your last letter, describing her “friend,” proves to me beyond doubt that you were ignorant of nothing but his name, for Greywood and no other answers exactly to that description. How you found out what you did, I can’t imagine; but remembering that your office window commands a view of the entrance to the building, I think it possible that you may have seen something from that point of vantage, which enabled you to put two and two together. But I wonder that I can feel any surprise at your having discovered the truth, when the truth itself is unbelievable!! May Derwent is incapable of killing any one—no matter what provocation she may have had. She is incapable of a dishonourable action, and above all things incapable of an intrigue. She is purity itself. I swear it. And yet what are the facts that confront us? A man, known to have been her professed suitor, is found dead in a room adjoining her apartment, dead with a wound through his heart—a wound, too, caused by a knitting-needle or hat-pin, as you yourself testified! And before trying to find out who killed him we must first think of some reasonable excuse for his having been at the Rosemere at all. How strange that he should happen to go to the building at the very time when May (who was supposed to be on her way to Bar Harbor, mind you!) was there also. Who was he calling on, if not on her? Luckily, no one as yet seems to have thought of her in connection with Greywood’s death. My sister has, in fact, been wondering all day whom he could have been visiting when he met his tragic fate. But, sooner or later, the truth will become known, and then—? Even in imagination I can’t face that possibility.
- 87. And now, since you have discovered so much, and as I believe you to be as anxious as I am to help this poor girl, I am going to accede to your request and tell you all that I have been able to find out about the sad affair. I know that I run the risk of being misunderstood—even by you—and accused of unpardonable indiscretion. But it seems to me that in a case like this no ordinary rules hold good, and that in order to preserve a secret, one has sometimes to violate a confidence. I have discovered—but I had better begin at the beginning, and tell you as accurately and circumstantially as possible how the following facts became known to me, so that you may be better able to judge of their value. Truth, after all, is no marble goddess, unchangeable, immovable, but a very chameleon taking the colour of her surroundings. A detached sentence, for instance, may mean a hundred things according to the when, where, and how of its utterance. But enough of apologies—Qui s’excuse, s’accuse. So here goes. I spent the morning on our piazza, and as I lay there, listening to the faint strains of familiar hymns which floated to me through the open windows of our village church, I could not help thinking that those peaceful sounds made a strange accompaniment to my gloomy and distracted thoughts. I longed to see May and judge for myself how things stood with her. I was therefore especially glad after the service was over to see Mrs. Derwent turn in at our gate. She often drops in on her way from church to chat a few minutes with my mother. But I soon became convinced that the real object of her visit to-day was to see me. Why, I could not guess. The dear lady, usually so calm and dignified, positively fidgeted, and several times forgot what she was saying, and remained for a minute or so with her large eyes fastened silently upon me, till, noticing my embarrassment, she recovered herself with a start and plunged into a new topic of conversation. At last my mother, feeling herself de trop, made some excuse, and went into the house. But even then Mrs. Derwent did not immediately speak, but sat nervously clasping and unclasping her long, narrow hands. “Fred,” she said at last, “I have known you ever since you were a little boy, and as I am in great trouble I have come to you, hoping that you will be able to help me.” “Dear Mrs. Derwent, you know there is nothing I would not do for you and yours,” I replied. “It is May that I want to speak to you about; she is really very ill, I fear.” “Indeed, I am sorry to hear it; what is the matter with her?” “I don’t know. She has not been herself for some time.”
- 88. “So I hear. Do you know of any reason for her ill health?” “She has not been exactly ill,” she explained, “only out of sorts. Yes, I’m afraid I do know why she has changed so lately.” “Really,” I exclaimed, much interested. “Yes, it has all been so unfortunate,” she continued. “You know how much admiration May received last winter; she had several excellent offers, any one of which I should have been perfectly willing to have her accept. Naturally, I am not anxious to have her marry, at least not yet; for when my child leaves me, what is there left for me in life? Still, one cannot think of that, and if she had chosen a possible person I should gladly have given my consent. But the only one she seemed to fancy was a most objectionable young man, an artist; the Maurice Greywood, in fact, of whose supposed murder you no doubt read in this morning’s paper.” “Yes,” I admitted. “Well, I put my foot down on that. I told her she would break my heart if she persisted in marrying the fellow. It was really a shock to me to find that a daughter of mine had so little discrimination as even to like such a person; but she is young and romantic, and the creature is handsome, and clever in a Brummagem way. The man is a fakir, a poseur! I even suspect, Fred, that his admiration for May is not quite disinterested, and that he has a very keen eye to her supposed bank account.” “But May is such a lovely girl——” “Oh, yes. I know all about that,” interrupted Mrs. Derwent, “but in this case ‘les beaux yeux de la cassette’ count for something, I am sure. He has absolutely no means of his own, and a profession which may keep him in gloves and cigarettes. I hear that he is supported by his mother and friends. Think of it! No, no, I could not bear her to marry that sort of man. But the child, for she is little more, took my refusal much to heart, fancied herself a martyr no doubt, and grew so pale and thin that I consulted the doctor here about her. He suggested nervous prostration, due to too much excitement, and wanted her to take a rest cure. I am sure, however, that that is all nonsense. May was simply fretting herself sick; she wanted to be ill, I think, so as to punish me for my obduracy.” “But what, then, makes you so anxious about her now?” I inquired. “Have any new symptoms developed?” “Yes,” and after glancing anxiously about to see whether she could be overheard, Mrs. Derwent continued in a lower voice. “You know that she started to go to Bar Harbor last Tuesday.” I nodded. “Well, she seemed really looking forward
- 89. to her visit, and when she left home was very affectionate to me, and more like her old self than she had been for months. But through some carelessness she missed her connection in town, and instead of returning here as she ought to have done, spent two nights in our empty apartment—of all places!! What possessed her to do such a thing I cannot find out, and she is at present so extremely excitable that I do not dare to insist on an explanation. When she did return here on Thursday she told me at once about the murder and how she was made to look at the body and to give an account of herself. Of course, we were very much afraid that her name would get into the papers and all the facts of her escapade become known. Through some miracle, that at least has been spared me; but the shock of being brought into such close contact with a mysterious crime has proved too much for the child’s nerves, and she is in such an overwrought hysterical condition that I am seriously alarmed about her. I wanted to send again for Dr. Bertrand. He is not very brilliant, but I thought he might at least give her a soothing draught. She wept bitterly, however, at the bare idea—insisted that he only made her more nervous. I then suggested sending for our New York physician, but she became quite violent. Really I could hardly recognise May, she was so——so—impossible. Of course she is ill, and I now fear seriously so.” Mrs. Derwent paused to wipe her eyes. “When you say that she is violent and impossible, what do you mean, exactly?” “It is difficult to give you an idea of how she has been behaving, Fred, but here is an instance that may show how extraordinary her conduct has been: Her room is next to mine, and since her return from town she has shut herself up there quite early every evening. I know she doesn’t sleep much, for I hear her moving about all night long. When I have gone to her door, however, and asked her what was the matter, she has answered me quite curtly, and refused to let me in. She has not been out of the house since she came back, but, strangely enough, I have caught her again and again peering through the blinds of those rooms that have a view of the road, just as if she were watching for somebody. As soon as she sees that she is observed, she frowns and moves away. Last night I slept very heavily, being completely worn out by all this anxiety, and was suddenly awakened by a piercing shriek. I rushed into May’s room and found her sitting up in bed talking volubly, while about her all the lights were blazing. ‘Take him away, take him away!’ she kept repeating, and then she wailed: ‘Oh, he’s dead, he’s dead!’ I saw at once that she was asleep and tried to rouse her, but it was some time before I succeeded in doing so. I told her she had been dreaming, but she showed no curiosity as to what she might have been saying, only evincing a strong desire to be left alone. As I was leaving the room, I noticed that the key-hole had been carefully stopped up. I suppose she did that so as to prevent my knowing that she kept her lights burning all night. But why make a secret of it? That is what I can’t understand! She has had a shock, and it has probably made her afraid of the dark,
- 90. which she has never been before, and perhaps she looks upon it as a weakness to be ashamed of. Another unfortunate thing occurred this morning. May has lately been breakfasting in bed, but, as ill-luck would have it, to-day she got down-stairs before I did, and was already looking over the newspaper when I came into the room. Suddenly she started up, her eyes wild with terror, and then with a low cry fell fainting to the floor. “Snatching up the paper to see what could have caused her such agitation, I was horrified to read that the man who was found murdered in our apartment house was now supposed to be Maurice Greywood. Imagine my feelings! As soon as she had recovered sufficiently to be questioned, I begged her to confide in me —her mother. But she assured me that she had told me everything, and that the man who had been killed was a perfect stranger to her and not Mr. Greywood. She insists that the two do not even look very much alike, as the deceased is much larger, coarser, and darker than the young artist. It was, of course, the greatest relief to know this. Had Greywood really been at the Rosemere on the evening she spent there, I should always have believed that they had met by appointment. ‘Yes, I should; I know I should,’ she repeated, as I shook my head in dissent. “When I was ready to go to church, I was astonished to find May waiting for me in the hall. She was perfectly composed, but a crimson spot burned in either cheek and her eyes were unnaturally bright. I noticed, also, that she had taken great pains with her appearance, and had put on one of her prettiest dresses. I could not account in any way for the change in her behaviour. As we neared the village, she almost took my breath away by begging me to telegraph to Mr. Norman to ask him to come and stay with us! ‘Telegraph him now!’ I exclaimed. ‘Yes,’ she replied; ‘I would like to see him. If we telegraph immediately, he could get here by five o’clock.’ ‘But why this hurry?’ I asked. She flushed angrily, and kept repeating: ‘I want to see him.’ ‘But, my child,’ I remonstrated, ‘I don’t even know where Mr. Norman is. He certainly is not in town at this time of the year.’ ‘Telegraph to his town address, anyhow, and if he isn’t there it doesn’t matter,’ she urged.—‘But, May, what is the meaning of this change? The last time he came down here you wouldn’t even see him. Do you now mean to encourage him?’ ‘No, no,’ she asserted. ‘Then I shall certainly not send him such a crazy message,’ I said. ‘If you don’t, I will,’ she insisted. We were now opposite the post office. She stopped and I saw that she was trembling, and that her eyes were full of tears. ‘My darling,’ I begged her, ‘tell me the meaning of all this?’ ‘I wish to see Mr. Norman,’ is all she would say. Now, I suppose you will think me very weak, but I sent that telegram. Fred, tell me, do you think the child is going insane?” and the poor mother burst into tears. “Dear, dear lady, I am sure you are unnecessarily alarmed. If I could see May, I could judge better.”
- 91. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com