THE DEVELOPMENT OF A REMOTE HELD
DATA ACQUlSlTiON SYSTEM

Thesis for the Degree of M. S.
MlcmGAN STATE UNIVERSWY
THOMAS WAYNE BOUCHER
1 9 7 6

 

 

J _‘ ' . 1 ' 7‘

fi ‘ ‘

_ .
‘. BINDING IV

: uunnsm [ 3

 

 

 

finer
ope:

char

are
ing
1.03:.

are

30m

ABSTRACT

THE DEVELOPMENT OF A REMOTE FIELD
DATA ACQUISITION SYSTEM

by

Thomas Wayne Boucher

This work examines the characteristics of a computer compatible
remote field data acquisition system. Both software and hardware char-
acteristics are considered. The work then presents construction details
and operational characteristics of a remote field data acquisition
system developed for the Department of Entomology at Michigan State
University.

The development indicates the flexibility available to the user
when recording data in a computer compatible format. Sensors can be
operated with a minimal amount of additional circuitry as nonlinear
characteristics can be linearized by computer manipulation of the data.

The environmental parameters of temperature and relative humidity
are measured and recorded by the developed system. The signal condition-
ing and sensors for the above measurements are described in this work.
Low-power complementary-symmetry metal over silicon-integrated circuits
are extensively used in the relative humidity circuits.

The system developed is but an elementary stepping stone to even

more complex systems.

THE DEVELOPMENT OF A REMOTE FIELD
DATA ACQUISITION SYSTEM
By

Thomas Wayne Boucher

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Electrical Engineering and Systems Science

1976

has 1
1 W01
Dr. '
men:

rize

r:
B
"(.1

wish

Beat

Last

Lim

ACKNOWLEDGMENTS

In the course of the production of this work, Dr. P. David Fisher
has provided much constructive criticism and invaluable encouragement.
I would like to express my most sincere appreciation for the aid which
Dr. Fisher has so unselfishly given. Dr. Dean L. Haynes of the Depart-
ment of Entomology has offered many helpful suggestions and has autho-
rized the dispensation of funds necessary for the purchase of the various
components used in the system development. For his kind assistance I
wish to express my most grateful thanks. I also wish to thank Mrs. Cindy
Beard for the competent work she has done while typing the final draft.
Lastly, I would like to express my sincere gratitude and love to my wife,

Linda, who has been so patient with me over these last months.

11

Chapter

II

III

TABLE OF CONTENTS

LIST OF TABLES. . . . . . . . . . . . .
LIST OF FIGURES . . . . . . . .
INTRODUCTION. . . . . . . . . . . . . . . . . .

REQUIREMENTS OF THE AUTOMATED REMOTE DATA
ACQUISITION SYSTEPI. O O O O O O I I O O C

General Operational Procedure. . . . . . .
General System Specifications. . . .

Hardware Considerations. . . . . . . . .
Software Data Management Considerations. .

NNNN
bWNH

SYSTEM IMPLEMENTATION . . . . . . . . . . . .

3.1 Introduction . . . . . . . . . . . . . . .
3.2 Signals, Sensors and Signal Conditioning .
3.2.1 Temperature Measurements. . . . . .
3.2.2 Humidity Measurement.

The Data Logger. . . . . . . . . . . .

The Reader System. . . . .

Sof ware Considerations. .

3
3
3

Uil-‘UJ

CIRCUIT RESPONSES AND TEST RESULTS. . . . . .
4.1 Thermistor Circuit Response. . . .

4.2 Humidity Sensor Circuit Responses. .
4.3 Data Logger Calibration. . . . . . . . .
CONCLUSIONS AND RECOMMENDATIONS . . . . . . . .
APPENDIX A: LISTING OF SUBROUTINE "INTRO". . .

APPENDIX B: EXAMPLE ILLUSTRATING THE EXECUTION
OF SUBROUTINE "INTRO" . . . . . . . . . . .

APPENDIX C: PROGRAM OCTLHDR.
APPENDIX D: PROGRAM OMVC . . . . . . . . . . .
APPENDIX E: PROGRAM TEMPLIN. . . . . . . . . .

iii

45

DOG-D

18
19
21
3O
54
61
68
74
74
79
88
91

96

98

100

101

102

Chapter

V

APPENDIX F:

APPENDIX G:

BIBLIOGRAPHY.

PROGRAM HMDTY.

iv

105

II

III
IV

v

XI

XII

Table

II

III

IV

VIII

IX

XI

XII

LIST OF TABLES

Stanley House Criteria for Humanizing
Information Systems. . . . . . . . . .

Representative Environmental Data
Required . . . . . . . . . . . .

Selecting Instrumentation Transducers. . .
Thermistor Characteristics . . .
Thermistor Board Terminations. . . . . . .
(a) Thermistor Board #7 . . . . . . . . .
(b) Thermistor Board #8 . . . . . . . . .
Header Data Display Example. . . . . .

Header Data Example. . . . . . . . . . . .

Examples of Computer-Stored Tables .

Bridge Output as a Function of Temperature .

Resistance, Bridge Output Voltage, Range
and Relative Humidity Tabulations. . .

Test Data Recorded on Digital Cassette
Tape by Incre Data "Incre-Logger" Model
5822 X/N 123 o o o o o o o o o o o o 0

Comparison of the Input Voltage to the
Recorded Voltage of the Incre-Logger .

10
12
22
32
32
33
62
71
73

80

84

89

90

Figure

10
ll
12
13
14

15

16
17
18
19

20

LIST OF FIGURES

General Data Collection and Management
Procedure 0 O O O O I O O O 0 O 0

Detailed Data Management Procedure. . . . . . .
Typical Physical Transducers. . . . . . . . . .
Signal Conditioning Unit. . . . . . . . . . .

Resistance vs. Temperature Characteristics
of "OMEGA" Thermistor . . . . . . . . . . . . .

Fundamental Thermistor Bridge Network . . . . .
Electrical Topology of Thermistor Board . . . .

Plot of Sensor Resistance vs. Percent Relative
Humidity for PCRC-ll Humidity Sensor. . . . . .

Block Diagram of Humidity Measurement Circuit .
Detailed Circuit of Range Indicating Network. .
Details of Automatic Bridge . . . . . . . . . .
Oscillator and Power Supply . . . . . . . . . .
Precision Rectifier Circuit .

Operation of the Precision Rectifier. . . . . .

Differential Amplifier and Low Pass Filter
Ne two rk O O O O O O O O I O O O O O 0 O O O O 0

Reference Sources and Window Comparator . . . .
Entire Humidity Measurement Circuit . . . . . .
Exterior and Interior View of Incre-Logger. . .
Block Diagram of Incre-Logger . . . . . . . . .
Photograph of Cassette Reader . . . . . . . . .

vi

16

20

20

23

23

31

35

37

39

4O

43

45

47

49

51

52

55

56

64

25

27

28

29

Figure Page

21 Photograph of Data Terminal and

Cassette Reader . . . . . . . . . . . . . . . . . . . . 66
22 Typical Data Terminal Printout. . . . . . . . . . . . . 67
23 Software Flowchart for Information Flow . . . . . . . . 69
24 Theoretical Thermistor Bridge Output as

a Function of Potentiometer a and 8 Factors

and Temperature . . . . . . . . . . . . . . . . . . . . 75
25 Graph of Theoretical Thermistor Output

Variations With Supply Voltage Variations . . . . . . . 77
26 Voltage Divider for Monitoring the

Power Supply. . . . . . . . . . . . . . . . . . . . . . 73
27 Graph of Measured Thermistor Bridge Response. . . . . . 83
28 Measured Humidity Circuit Response. . . . . . . . . . . 87
29 CMOS low-power clock. . . . . . . . . . . . . . . . . . 93

vii

the

VET

tic

tim

PIE

all

“P

rec

num;

Dad.

CHAPTER I

INTRODUCTION

Natural scientists often find that extensive time delays occur when
they attempt to acquire and process field data. Whether that entails
verifying a theory in the area of pest management or the simple collec-
tion of data for analysis of possible factors affecting healthy apple
growth, the traditional methods of data acquisition remain as time con—
suming, and often times frustrating, stumbling blocks. In the past,
collection and cataloguing of viable data has taken up the complete
lifetimes of many researchers. Due to this large degree of human inter-
action, development of useful ecosystem management strategies has been
impeded. Furthermore, these same time delays make the implementation
of these management strategies very difficult in many cases because real-
time climatological and biological data is not available concerning the
present state of the system being managed. The purpose of the research
program reported in this thesis is to develop and evaluate useful tech-
niques and methodologies which will substantially reduce these time
delays.

The conventional approach entails either (1) the researcher manu-
ally making his own real-time measurements in the field or (2) he sets
up a measurement system in which desired data is recorded on strip-chart
recorders. The first method becomes progressively prohibitive as the
number of observations increases or if a particular observation must be
made continuously over extended periods of time. Not only that, but

1

the <
cone:
in t]
The '
the
one'
The
plac
I read
ing
the
UP or

P€at

hon}
Sit:
PUII
ins-
eHal

ent;

the time it takes to transcribe the data from an instrument to a note-
book balloons correspondingly. Also, due to the nature of the record-
ing medium, extensive time is required when the scientist converts the
analog voltage output recorded on the chart paper to the corresponding
parameter he attempts to measure and then again transcribes this into a
computer-compatible form for purposes of data storage and/or analysis.

Consider the following example. A physiographer (one who studies
the descriptions of the features and phenomena of nature) desires data
concerning soil temperature at various soil depths. He is interested
in the variation over a time period covering fall, winter and spring.
The temperature variation at depths of 10, 20, and 40 centimeters is
the scientist's major concern. He contacts one of the local growers --
one who has offered 60 acres of his fallow ground for use in the project.
The physiographer then constructs three test sites. On each site, he
places thermocouples at each of the three selected depths. Temperature
readings are taken every three hours. The methodology consists of tak-
ing a potentiometer bridge out to each test site and connecting it to
the thermocouple leads. The temperatures at each depth are recorded
upon a chart prepared for the corresponding site. The process is re-
peated every three hours for the next nine months.

After several months, the physiographer realizes excessive man-
hours have been consumed in the course of implementing the data acqui-
sition phase of this study. He therefore executes the following changes.
Purchasing nine single—track strip-chart recorders, he has workmen
install them at the test locations. The sensitivity of the recorders
enables direct connection of the thermocouples. Additional equipment

entails hookup leads and environmental shelters for the chart recorders.

At the
of cha
ing WI

fortir

must 2
UIEDII

suits

data.
catal
ful c
data

verif
and c
requj
this

resu:
VidEI

tial

At the end of each week, a research assistant brings out new supplies
of chart paper and verifies proper operation of the equipment. Return-
ing with the collected data, he sits down to spend the next week trans-
forming the recorded voltages into corresponding temperatures.

The transform process can be very lengthy and during the trans-
cription mistakes may easily occur. After transformation, the data
must again be plotted in order to easily recognize and verify trends
within the data. Again, extensive human manipulation of the data re—
sults in long time delays and greater probability of error.

This paper describes a tool to aid in the systematic collection of
data. It minimizes the time spent by researchers when collecting,
cataloguing and transforming raw data. The tool incorporates the power—
ful computational abilities of a computer, which aids the researcher in
data analysis. It allows one to quickly discern new relationships and
verify old ones in any data base. An overview of the data acquisition
and data managment system is provided in Chapter II, along with the
required specifications for the system. The hardware implementation of
this system and the software is given in Chapter III. Preliminary
results of evaluating this automated data acquisition system are pro-
vided in Chapter IV. A summary of the results and a list of the poten-

tial uses and limitations of this system are given in Chapter V.

when
syst
basi
Iogg
tion
desir
opera
coast
in Se
hard;

Speci

2.1

user |
the c.
deter:
the SE
data I
fOrms

wbile

CHAPTER II
REQUIREMENTS OF THE AUTOMATED
REMOTE DATA ACQUISITION SYSTEM
This chapter escablishes the criteria which must be considered
when developing a remote, automated data acquisition and data-logging
system, as well as the concomitant data management systemJ‘Also, the
basic procedure to be followed when using an automated remote data
logging system is described. The basic operations of the data collec-
tion process are formulated in Section 2.1. Also in this section, the
desired user operational procedures are generalized. Once the desired
operational format has been established, the general specifications and
constraints remain to be determined. These specifications are described
in Section 2.2. After establishing the above, Section 2.3 details the
hardware specifications which must be met and Section 2.4 analyzes any

specifications which the software must meet.

2.1 General Operational Procedure

 

The operational procedure should be easily followed by the specific
user of the system. Figure l flowcharts the typical procedure used in
the collection of data via automated data-logging techniques. The user
determines the type of data he requires. He examines the properties of
the sensors available to determine those which will monitor the desired
data more effectively. If laboratory calibration is required, he per-
forms such calibration before positioning his sensors at the field site.
While positioning the sensors, the researcher locates a central field

4

 

 

Problem
Identification

 

 

 

 

Reach
Conclusions

 

 

 

 

Determine
Sensors
If Necessary

 

Search

for

Solution

 

 

 

 

.___.L___

Laboratory
Calibration
If Required

 

Researchertrans~
ports Logging
r-d Equipment To
Remote Test Site

 

 

 

_ I

 

 

 

 

 

Apply
Linearization,
Statistical
Algorithms
If Required

 

 

   
 

Perform
Analysis
Of Data

    

 

 

P-—_—q

 

 

Data May be Lef
In Storage For
Recall at User'
Convenience

 

Transcribe
Data Into
Permanent
File

 

Figure 1.

 

Position
logger and sen-
sors as desired

 

 

 

Data
Periodically
Sampled And
Recorded

 

 

 

Bulk Storage
Media
Renewed

 

 

 

 

 

.____i__1

Minor
Maintenance

 

 

—-+V

 

 

Reset Logger,
Return To
Laboratory With

Recorded Data

 

 

Performed

 

 

As Required

 

General Data Collection and Management Procedure

site where he places the recorder. After concluding that the system is
as desired, he returns to the office. The system user determines either
that sufficient data has been recorded or that the recording mechanism
is in need of maintenance and replacement of the bulk storage media.
Returning to the field site, he/she retrieves the already recorded
data from bulk storage and any minor maintenance is performed as re-
quired. The media upon which data is stored is replaced, whereupon the
researcher proceeds back to the laboratory with data in hand. Immedi-
ately upon return, the user logs onto a main computer system and trans-
fers the data from the bulk storage media into a permanent file. At
the user's convenience, the permanent file is attached and data analysis
is implemented as required or desired. The user then determines if more
data would aid analysis. If deciding affirmatively, he continues to
allow the system to operate as described above; otherwise, a return to
the test site is made and the equipment is dismantled or moved to
another location. A thorough inquiry of the system specifications and
allowable constraints must be undertaken to adequately implement the

above procedure.

2.2 General System Specifications

 

The primary users of this system will be natural scientists. The
apparatus must therefore be designed such that the researcher obtains
maximum satisfaction with a minimum of user frustrations. Theodore D.
Sterling [1] has compiled the table which is reproduced in Table I, which
sets forth many of the criteria for humanizing a computerized informa-
tion system. His table applies specifically to software systems and,
as such, must be carefully considered when structuring the data manage-

ment portion of this work. However, many of the points covered in the

TABLE I

Stanley House Criteria for Humanizing Information Systems

A. Procedures for dealing with users

1. The language of a system shouldlneeasy to understand.

2. Transactions with a system should be courteous.

3. A system should be quick to react. .

4. A system should respond quickly to users (if it is unable to resolve
its intended procedure).

5. A system should relieve the users of unnecessary chores.

6. A system should provide for human information interface.

7. A system should include provisions for corrections.

8. Management should be held responsible for mismanagement.

B. Procedures for dealing with exceptions

1. A system should recognize as much as possible that it deals with
different classes of individuals.

2. A system should recognize that special conditions might occur that

could require special actions by it.

A system must allow for alternatives in input and processing.

A system should give individuals choices on how to deal with it.

A procedure must exist to override the system.

LDJ-‘U
.00

C. Action of the system with respect to information

1. There should be provisions to permit individuals to inspect infor-
mation about themselves.

2. There should be provisions to correct errors.

3. There should be provisions for evaluating information stored in the
system.

4. There should be provisions for individuals to add information that
they consider important.

5. It should be made known in general what information is stored in
systems and what use will be made of that information.

D. The problem of privacy

1. In the design of a system all procedures should be evaluated with
respect to both privacy and humanization requirements.

2. The decision to merge information from different files and systems
should never occur automatically. Whenever information from one
file is made available to another file, it should be examined first
for its implications for privacy and humanization.

E. Guidelines for system design having a bearing on ethics

1. A system should not trick or deceive.

2. A system should assist participants and users and not manipulate
them.

3. A system should not eliminate opportunities for employment without
a careful examination of consequences to other available jobs.

4. System designers should not participate in the creation or mainte-
nance of secret data banks.

5. A system should treat with consideration all individuals who come
in contact with it.

tabl

acqu

U581

UDDE

of

The

rug

of

pr

Vi

aI

table can and should be reapplied when developing the hardware of data
acquisition systems. The hardware should be constructed so that the
user can easily set up and operate it. It should relieve the user of
unnecessary hardware maintenance. It should include provisions for
easy interfacing with various transducers. It should provide the abil-
ity to easily increase the number of data channels. In terms of user
convenience, the entire recording system should be relatively compact.
One person should be able to install it easily in any location. As
such, the system must be portable. Its operation must be independent
of the power line system for ease of installation in isolated areas.
The power source must also be portable and able to provide operating
power for a considerable length of time. In consultations with natural
scientists, it is agreed that an ideal minimum time span of uninter-
rupted and unattended operation would be one month.

In addition to the harsh power requirement, climate conditions
which the system may be subject to must be taken into consideration.
System operation should cover an ideal temperature span of -50°C to
+50° c or -58°F to +122°F with very little degradation in data handling
ability or in accuracy of the recording process. The temperature range
of operation of the system is just a singular aspect of a much greater
problem. The recording units and sensors must either be capable of
withstanding or be protected from the corrosive effects of humidity,
condensation, and air borne chemical pollutants. Protection of the
equipment from the weathering of wind blown sand, snow and/or ice must
also be provided.

Weather is not the only culprit contributing to the destruction

of viable data. Many small animals can wreak havoc with interconnecting

5531

small

osit

syst

tIo:

£101

Us

13111

COIL

(\J
o
(A)

and
t6]
cor
II
hoi
til
Qu.
ti:
Pe

Ch.

cables, sensors and quite possibly the recording instrument itself.
Small animal instincts drive the animal to fashion nesting areas within
small spaces contained in any equipment, and many exhibit extreme curi-
osity concerning the taste of various sensors and cables. Thus, the
system physically must be quite durable.

In addition, since the recorded data provides input for a computer-
based data management system, the recording process and media should be
directly compatible with the computer. The computer processes informa-
tion in the form of electrical signals which represent bits of informa-
tion. Any compatible recording process must then be capable of
transforming analog data signals into corresponding digital bits. The
bulk storage must then be capable of directly interfacing with the
computer. Some form of digital-to-analog conversion must be chosen to
realize the desired transform and the transformed data must then be

stored on some form of digital recording media.

2.3 Hardware Considerations

 

Before determining the hardware specifications which the sensors
and the recorder must meet, it is advantageous to examine some charac-
teristics of the data to be preserved. Most natural scientists are
concerned with the representative environmental data listed in Table
II [2]. For each of the various parameters, the range, accuracy, thres—
hold value, and the maximum rate of change of the parameter per unit of
time are specified. The fastest rate of change is 30 mph/sec. This
quality of the wind speed is a limiting factor placed upon the response
time of the recording system. If the recorder operation is such that
periodic samples of the parameters are recorded, then the high rate of

change of wind speed constrains the system to sample at a rate of twice

10

 

 

u--- i--- Now aneanduum oa-o H ewes
III: 0 w\~Eo\Hmo m+ v\~8o\Hmo oomuo H 30H
I. . auHmcoocH ustH
L\Nmm No NNH+ muHomamo oHon NOOHno H ousumHoa HHom
n\.sH omumH .cH H.o .cH Hum HHmHaHmu .cH «no H :OHomuHaHooum
E\No~ Na NHHU Nmmuo H sanHaam
a\wm .ae mm.o we .aa AN mm .aa ~.o+ we .ae m.Hmuo.a~ H teammate denudeonem
aae\ooma ans H ooam ooemuo H coauoanHe was;
ddn\nae on gas H and H+ sea onto H emeam seas
AoommusmlnsmV £\ooH I.
Addameanv aHs\ooH u--- UOHH HmoomHimwv uoemumu A.aa m.m.H.oV a anaenndaade Haom
cHE\ooH nun: ooH+ Amoonummlv oomcumqu H monumuoaaos uH<
mafia mo uwca cHosmouafi xommsoo< owmmm ouHm Hon mommamumm
mom mouoamumm poHHsvmu
ozu mo owcmso mumoaonsmmoa
mo mums adstmz mo monasz

 

HH mHm<H

HNH voHHsvom some HmusoscooHpam o>HuMucomooaom

a 2:
pre

mi

of

tic
at}
hax

the

tat
is

mer
DEC
PEI
Sig
Val

C0!

0pe
duc
tul
of 1

Si:

11

a minute if parameter changes are to be accurately followed. At the
present time, wind speed is not a critical measurement parameter in
this application. Initial system feasibility shall be demonstrated via
measurement of both soil and air temperatures in conjunction with that
of relative humidity. However, system design must take into considera—
tion the possible future addition of wind speed transducers and that of
other parameter sensors. Once the characteristics of the measurands
have been laid down, a foundation exists which aids in the selection of
the proper transducers to accomplish the measurement goals.

Table III pinpoints the important aspects of selecting instrumen-
tation transducers[3]. It is quite possible that a specific transducer
is chosen, but its input/output characteristics do not fit the require-
ments of the application, or it is found that signal amplification is
necessary. When additional electrical processing or amplification is
performed on a transducer output variable, the techniques are known as
signal conditioning. Signal conditioning makes the recording of small
variations in low level transducers possible. At the same time, it may
compensate for any non-linearities in the transducer output.

The components of signal conditioning circuits must be able to
operate throughout the same environmental temperature span as the trans-
ducers. In many cases, the signal conditioner itself provides tempera-
ture stabilization of the transducer output. It may be designed to
offset zero shift in a transducer by superimposing an equal, but oppo-
site, signal change on the transducer output.

When the transducer and its associated signal conditioning are
connected together, it is of the utmost importance that one consider

the effects of extraneous signals which could enter into the components.

12

TABLE III

Selecting Instrumentation Transducers

 

Characteristics of the Input Variable

Range (maximum and minimum values to be measured)
Overload protection
Frequency response

Transient response

Resonant frequency

Transducer Input/Output Relation

Accuracy
Repeatability
Linearity
Sensitivity
Resolution
Friction
Hysteresis/backlash
Threshold/noise level
Stability

Zero drift

Loss of calibration with time

Overall System

Output characteristics

Size and weight

Power requirements

Accessories needed

Mounting requirements

Environment of transducer location

Crosstalk

Effect of presence of transducer on measured quantity
Need for corrections dependent on other transducers

Measurement Reliability

Ease and speed of calibrating and testing

Time available for calibration prior to and/or during use

Duration of mission

Stability against drift of zero point and proportionality constant

Vulnerability to sudden failure (probability of proper performance
for a given life time)

Fail safety (will transducer failure represent system failure or
invalidate data from other transducers)

Failure recognition (will transducer failure be immediately apparent
so that subsequent erroneous data can be rejected)

Purchase

Availability and delivery; is item off-the-shelf?

Any development necessary for operation

Availability of calibration and test data from manufacturer
Price

Previous experience with seller

‘

uatec
and/c

CUSSJ

trans
be ta
able
inpuI
trans
POIaI
mate
The 1
a fo:
type

Side:

Verti
main1
IIOn

Anal,

13

A most frequently encountered source of error is the introduction of a
60-cycle power line signal. This signal can easily superimpose upon
the transducer signal via the cable interconnect between the transducer
and the signal conditioning circuits. In applications where do analog
voltages are measured, the 60-cyc1e component can be effectively atten-
uated by low-pass filtering, establishment of well-defined grounds,
and/or implementation of electrostatic shielding. For a thorough dis-
cussion, see Chapter 17 of Practical Instrumentation Transducers [3].
Many problems are encountered in the process of connecting the
transducer with the recorder. The output range of the transducer must
be tailored to the input range of the recorder. The recorder should be
able to withstand signals many orders of magnitude in excess of its
input range, with little or no damage to its components. Since many
transducers output a differential signal, the recorder should have bi-
polar inputs. The recorder must operate over the same extremes of cli—
mate conditions, under the same criteria of minimum power consumption.
The recorder, as already noted, must be computer compatible; as such,
a form of analog-to-digital conversion must be decided upon. The code
type software manipulations can easily perform conversions. Prior con-
sideration of the signals to be recorded shows that conversion speed is
not a necessity. In the conversion process, errors may occur due to
offset factors, gain factors and nonlinearities inherent in the con-
verter. The converter should be designed to minimize these errors and
maintain the minimal error over the entire span of temperature opera-
tion. A good discussion of analog-to-digital converters is given in

Analog Devices' AnalogrDigital Conversion Handbook [4].

 

soul
I

the

DEED

ftl]

inte

UPDI

O.
a
-1

fl

quar
pove
data
medf
varj

date

2.4

Tab]
Shot
Plea
the

Uhde

14

The system must process and convert several data channels. Eco—
nomic factors indicate that multiplexing the various data channels into
the A/D converter is desirable in terms of reduction in hardware com-
onents; thus, costs are lowered and portability increased.

A method of lowering power consumption, which has been used success-
fully, operates upon the principle of sampling the data at predetermined
intervals and disengaging power to the system between sampling times.
Again, the signals of concern vary slowly enough that the application
of the above principle increases the operating life of the system, with
no degradation of information. To implement the technique, some sort
of timing mechanism must be incorporated into the system.

There remains one facet of the system which has not been touched
upon. The hardware which provides the digital coding must be augmented
by a storage media for the digital codes. A mechanism to store bulk
quantities of data must be considered. The mechanism must exhibit low
power dissipation, maintain portability criteria and promote ease of
data collection while storing voluminous amountScxfdata. The storage
media should maintain data integrity over long periods of time and in
various environmental conditions. The media should be such that quick
data transfer from the media to a mainframe computer is possible.

2.4 Software Data Managgment Considerations

Earlier in this chapter a reference was made to the article on
"Humanizing Computerized Information Systems" [1]. Again, consider
Table I from this article. Many of the guidelines proposed in the table
should be followed if the software is to properly and efficiently im-
plement user interactive data management. In a data management system,

the language used should be both clear and easy for the operator to

understand. The system should quickly react to any requests or to

info
shou
cord
diff
stru
well
shoe
data
imp]
soft
EitI
cons
free
hove
acci

able

The
the
Phel
Pril
Bro:
the
inv;
0nc«

15

inform of system inability to meet the request requirements. The system
should eliminate unnecessary or redundant requests while providing
cordial human information interfacing. The ability to deal with many
different classes of individuals is a basic system criterion. The
structure should be able to satisfy both the computer-oriented user as
well as the researcher with little computer background. The system
should be structured to allow alternatives both in input procedures and
data processes. The software should be structured so as to efficiently
implement the application of currently existing hardware. However, the
software should be flexible in order to easily update the entire system
with improved hardware developments. Another aspect which deserves
consideration is that of data bank protection. The system should allow
free access of all data to the user. The presentation of that data,
however, should be accomplished in such a manner that the user does not
accidentally destroy it. At the same time, authorized users should be
able to quickly add to sections of the already catalogued data base.
Figure 2 illustrates typical user interactive data management.
The user obtains the bulk storage media with the accumulated data from
the field site. He reads this data, via a computer controlled peri-
pheral, and it is placed into temporary storage. The user then requests
printout of several lines of raw data. This enables him to eliminate
gross malfunction of the logger system as a source of error, and at
the same time, verify valid data transfer. If, at this point in time,
invalid data appears, it may be easily deleted from the data base.
Once satisfied that the data has been read and stored properly, the
user then catalogues it permanently in the form of a permanent file.

In effect, this negates the possibility of losing the valid data.

l6

 

 
  

  
  
  
 

 

 

 

 

 

 

 

   
 

 

 

 

 

 

 

 

 

 

Recorded
2:6:age -------- .~ Field
Site
Data
Transit From
_________ ‘_d Field To
‘ Computer
Terminal
Data Read By
Computer
Reading ----- 1 Controlled
Peripheral
I
l
I
l
Manual
Machine Override If
Storage Desired
Gross Error
Error 4 Editing If
Checking Necessary

 

 

 

 

 

Mathematical And

 

 

 

 

 

   
 

  

 

 
    

Data Statistical
Catalogued Algorithms
May Be Applied
I
l
I
I
Data ‘ Data
Linearization Data Plots
If Required Transformations if
Desired

    
  
 

 

 

  

Results

Figure 2. Detailed Data.Management Procedure.
Catalogued

  

Eurt

ehic
orig

USES

the
data
then

fer]

17

Further data processing may be halted at this point until a later time
or the user may directly proceed as follows.

He has the computer input the data into a linearization algorithm
which compensates for the measured nonlinearities of the sensor which
originally produced the signal recorded by the data logger. He then
uses the computertxatransform the linearized data into the corresponding
parameter which the sensor measures (i.e., temperature. dew point,
relative humidity, etc.). Using the computer, he tabulates the para-
meter values versus time and then plots a graph of it. Scrutinizing
the data, he decides to apply a statistical algorithm to two of the
data channels in hopes of discovering if a correlation exists between
them. After this analysis, the results are catalogued for future re-
ferral by either himself or his colleagues.

The above represents a general approach to an automated data-
management system. The realization of such a system develops in the

succeeding chapters of this work.

3.1

u:
u l
n.)
{-1

the

logs
fact
pane
than
to a
chat
the

ing

IEpz
Sect
Ehi]
Tecc
com;
no,
“hit

EEnI

CHAPTER III

SYSTEM IMPLEMENTATION

3.1 Introduction

 

The system, as developed for use by the entomologists at Michigan
State University, provides the topical material for this chapter. At
the time of the inception of this project, specifications of the data
loggers then available were considered. As a result, a logger manu-
factured by the "Incre-Data" Corporation was chosen as an ideal subcom—
ponentiinrrealizing the system. The logger model purchased has 20
channels for data input, and an input voltage range which is constrained
to a span of 100 mV covering the values of -50 mV to +50 mV. In this
chapter, Section 3.2 not only elaborates upon the characteristics of
the sensors, but also details the construction of the signal condition-
ing required. The signal conditioning provides an input voltage
representation of the data which falls within the above-mentioned range.
Section 3.3 fully develops the characteristics of the data logger,
while documenting the information needed to operate the logger. The
recovery of the data stored by the logger is facilitated by hardware
components. Section 3.4 lists the properties and describes the opera—
tion of those components. Section 3.5 tabulates the programs available
which form the basis for the software implementation of the Data Manage-

ment System.

18

able

the

indi

the:

dire

proc

a)
:1
CL

tect
erat

atte

is s
The
of .1
iUSt
nitr

Side

19

3.2 Signals, Sensors and Signal Conditioning
The photograph in Figure 3 shows some of the various sensors avail-
able which record geophysical data. The host of devices ranges from
the solar calorimeter in the background to the thermistor temperature
indicator in the foreground. Some of them are direct reading, as is the
thermometer, while others must be built into electrical circuitry, as
do the soil moisture sensors, in order that the signal may be measured
directly. Creation of sensors to measure various data is an on—line
process with new products constantly entering the marketplace.
Currently, our attention is focused upon those sensors which mea-
sure temperature and relative humidity. Table II in Chapter II lists
the signal characteristics along with the quality and quantity of the

measurements to be made. Frank J. Oliver, in Practical Instrumenta-

 

tion Transducers [3] discusses the merits and deficiencies of many of

 

these sensors.

Once the merits of the sensors were weighed, the actual design
and implementation of the signal conditioning circuits remained. Pro-
tection of the electronics from the environment being a prime consid-
eration, an aluminum casting, Figure 4, was purchased to house the
attendant hardware. Its dimensions (6 1/2" high x 6 1/2" wide x
10 3/4" long) facilitate its use in portable applications. The casting
is sealed by a large O-ring stretched around the perimeter of the lid.
The lid is then held solidly in place by six hex bolts. The protection
of interior components is insured by an air purging system. The valve
installed on the top of the lid makes it possible to introduce dry
nitrogen into the sealed casting, while a purge valve located on the

side of the unit bleeds off any moisture containing air previously

20

L,. ., Ll.~, "';’1*M “)1 >— i. , ~

‘>K||“
, ..A« tavWMVM mum”
IJ‘Mxlrle‘5‘il . u
~r‘Vy-mrihrvv,

"I

_- -. ....--.,.. -. ,-- -o---~--as-
h-~"o. . V, ;-—-.‘..‘ _—-o
.,-.-------1--l--. .- .. -.',-..

I '

 

Figure 3. Typical Physical Transducers

 

H ..)- Ah
.. «v .1»? e— »ma*'twrexm use"
‘ _ V Kw -\ \ , ' -‘ _.« . : . .

Lj)‘_“' wax .~._ -‘ ‘
yer) “IA ‘1 4W5}? ”3 \‘fi (T‘h‘i

., § 7 ." ... . t ,
~"fl "'H -" '3.Jv.'l."’.°‘.4l ‘ tit. - ', ‘V'

.,. s I’I - . - .,.,- s 3L... - . .»|"M"'Il ,...r._m.mm;'.x. v M‘ ‘ U . ‘. fifl‘ ‘ . . -
an. “VON-ca flown... . s_

'1”-_ __~-‘ '
“"‘ O
. .v

 

Figure 4. Signal Conditioning Unit

con
pos

the

the
sis
ori
the

PTO]

an
trat
acc‘

the

I811]!

Iv

IEmF

21

trapped. Along the sides of the box, near the corners, four male
connectors are mounted. Two of these connectors are 26 pin types
positionally located side by side. These have been chosen to carry
the conditioned signal to the logger and represent the outputs of the
signal conditioner. The other two connectors contain 36 pins apiece
and represent the sensor inputs to the unit. In Figure 4. the above-

mentioned features are pointed out.

3.2.1 Temperature Measurements

 

Thermistors were decided upon as the elements best suited to sense
the temperatures of both soils and air. These thermally dependent re-
sistors possess extremely high sensitivity and are able to hold their
original calibration for extended periods of time. Furthermore, unlike
thermocouples, no reference temperatures are required. High sensitivity
promotes their use in comparatively simple measuring circuits with
little degradation of accuracy due to changes in other components. As
an example, Oliver [3] states that 400 feet of No. 18 AWG copper wire
transmission line subjected to a 50°F temperature change affects the
accuracy of thermistor measurement by only 0.10F. This is well within
the accuracy range stated in Table II.

An "Omega Engineering" thermister has been chosen to perform the
temperature measurements. Its characteristics are tabulated in Table
IV and plotted in Figure 5. Its resistance variation as a function of

temperature can be approximated by the following formula.

8(
R(T) = Ro(To)e

l
T T )
(1)

22

TABLE IV

THERMISTOR CHARACTERISTICS

 

 

 

 

Temperature Resistance Temperature Resistance
(PC) (Kilohms) (0C) (Kilohms)
-30 88.5 - 6 22.31
-29 83.15 - 5 21.165
-28 78.2 - 4 20.08
-27 73.6 - 3 19.055
—26 69.25 - 2 18.095
—25 65.2 - 1 17.185
-24 61.4 0 16.325
-23 57.85 + 1 15.515
-22 54.55 + 2 14.75
-21 51.45 + 3 14.025
-20 48.535 + 4 13.345
-19 , 45.805 + 5 12.695
—18 43.245 + 6 12.085
-17 40.845 + 7 11.505
-16 38.59 + 8 10.96
-15 36.475 + 9 10.44

-14 34.49 +10 9.95
-13 32.62 +11 9.485
-12 30.865 +12 9.045
-11 29.215 +13 8.625
-10 27.665 +14 8.23

- 9 26.2 +15 7.855
- 8 24.825 +16 7.5

- 7 23.53 +17 7.16
+18 6.84 +35 3.265
+19 6.535 +36 3.1335
+20 6.245 +37 3.0085
+21 5.97 +38 2.8885
+22 5.71 +39 2.7735
+23 5.46 +40 2.6633
+24 5.225 +41 2.5585
+25 5.0 +42 2.4585
+26 4.7865 +43 2.3635
+27 4.5835 +44 2.2715
+28 4.3885 +45 2.185
+29 4.2035 +46 2.1005
+30 4.0285 +47 2.02
+31 3.8615 +48 1.945
+32 3.7015 +49 1.8715
+33 3.5485 +50 1.8015
+34 3.4035

Iigu

1.3V

23

RESISTANCE OHMS

 

 

 

0
Temperature C

Figure 5. ResiStance vs. Temperature Characteristics of "OMEG " Ther-

 

 

 

mistor.
0.10
‘—-——NV‘ ‘
fl 1961‘“ 1961(0
11 m
1 3V —_ 10m (10— VOUT+OTT 10m
T- a B\, fl
7009
IZKO SOKR (<f’T‘:l’ RT
3

 

 

Figure 6. Fundamental Thermistor Bridge Network

Rheat

tempt

Ether.

as f
tech
the

aVai
of c
mize
the

app]

in‘

24

R0(To) = Thermistor resistance at absolute temperature T0
= 5 x 103 ohms,
where To = 298.15°K = 25°C.
Ro(T) = Thermistor resistance at absolute temperature T
B = Thermistor material constant

3.82 x 104 Kelvins

Note: The value of B is determined from measurements made at 00C
and 50 C.

The thermistor is used in a circuit configuration known as the
Wheatstone bridge (see Figure 6). The output voltage as a function of

temperature can be expressed as:

 

(3.5 + B - a)RT + (a - a - 1.5)5 x 1040
vOUT = 1.3V 4 (2)
(29.6 + 1.78 + u)RT + (21.1 + 1.78 + a)5 x 10 Q
where
RT = 5 x 1039 exp 3.8 x 103(% - 1’ o ) . (3)
298.15 K

The component's values indicated in the schematic were arrived at
as follows: The 0.10 represents the internal resistance of the "Gould"
rechargeable nickel cadmium batteries. The 196 k9 resistors in two of
the legs of the bridge were chosen out of convenience as they were
available in abundance as a stock item. Their major function is that
of current limiting the bridge in order that power consumption be mini-
mized. The ISkO resistor in series with the 10 k9 potentiometer forms
the reference leg of the bridge. The reference value can be varied by
appropriate adjustment of the potentiometer. The variation is reflected

in the value of alpha which ranges from 0 to l. The other leg of the

bridge 1
nation a
earl is
the othe
:eter.
be 7003
roltage
DEHIS .
Wr

IN «I

(196%

25

bridge is composed of a 50 k9 resistor in parallel with a series combi-
nation of thermistor and the interconnecting cable. The parallel net—
work is placed in series with a 10 k9 potentiometer, thus completing
the other leg. Beta represents the variability of the lOkQ potentio-
meter. A nominal dc resistance for the coaxial cable was measured to
be 7009 for a 15 foot length. By applying mesh analysis, the output

voltage V can be determined in terms of the various circuit compo—

OUT
nents.

Writing the loop equations,

-l.3V + .1911 + 196k9(il—12) + 12kn(il-12) + 10k9(i1—12) = o, (4)

(l96kQ + leQ + lOkQ) (iz-il) + l96k§2i2 + 50kQ(12-i3) + lOin2 = 0 (5)

50k9(13-12) + 7009i3 + RT 13 = 0 (6)

Rearranging,the equations become:
218.001kOil - 218in2 = 1.3V, (4')

-218ktiil + 474in2 - 50in3 = 0, (5')

—50kfli + (50.7k9 + RT)i3 0. (6')

2

Using (6'), solve for i3 in terms of 12.

SOin2

3 = (50.7kQ + RT) (7)

 

i

Substituted into (5'), the equations reduce to:

 

(2.18001x105mi1 - (2.18x105mi2 = 1.3V (8)
P
5 2.15318x10lo + 4.74x105RT
-(2.l8x10 )5211 + 4 912 = o (9)
L 5.07x10 + RT

 

35

Th:

(:-
H

80

26

Equations (8) and (9) can be put into matrix form and evaluated

as follows:

"’ «I!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.0x105(2.18001) -1.0x105(2.l8) 31' [1.5
5 5 2.15318x105 + 4.74RT
-l.0x10 (2.18) 1.0x10 4 i2 0
5.07x10 + RT . J . J
L. .4
(10)
" W P . - I
2.18001 -2.18 11 1.3
1.0x105 5 =
2.15318x10 + 4.74RT
-2.18 4 12 0 (11)
5.07x10 +RT (l. J .. .Il.
The determinant of the coefficient matrix is:
5 2 (2.18001) (2.15318x105 + 4.74RT) 2
A = (1.0x10 ) 4 l (2.18) (12)
(5.07x10 + RT)
which can be reduced to the following form:
5 2 (2.284467753x105 + 5.58080474RT)
A = (1.0x10 ) 4 (l3)
(5.07x10 + RT)
Solving for 11 and 12 by Cramer's formula:
P 5 an
1.3 -l.0x10 (2.18)
5 2.15318x10S + 4.74RT
0 1.0x10 4
5.07x10 + RT
11 = , *- (14)
A
1.0x105(2.180001) 1.3
5
i = 7 -1.0xlO (2.18) 0 (15)
2 A

Equations (14) and (15) can be reduced to:

i b)

atta

beco

PIES

V
DUI

fun:
V
OU'

1.3

By

27

 

 

 

5
2.15318x10 + 4.74
11 = 1.3 5 5 RT (16)
1.0x10 L2.284467753x10 + 5.58080474RT
P 5
1.10526x10 + 2.18
1.0x10 L2.284467753x10 + 5.58080474RT

The power supplied by the battery may be calculated by multiplying

i by the supply voltage.

1
P = 1.311
5
_ 1.3 2 2.15318x10 x 4.74RT
_ _JL__JL_ 5 (18)

 

1.0x105 2.284467753x10 + 5.58080474RT

The greatest power dissipation occurs when the resistance of RT
attains a low value of 1.8015k9, then the maximum power dissipation
becomes 15.86uW.

The bridge output voltage, as a function of i1 and 12 can be ex-

pressed as:

4 (1.2 + a + 8) (5.07x104) + (6.2 + a + e) (700 + RT)

V = 1.0x10 1
OUT 5.07x104 + RT 2

 

 

- (1.2 + a)1 (19)

Substituting (16) and (17) into (19), the output voltage, as a
function of the circuit resistances, is obtained.

V =
OUT I

(7.828+2.18B-2.56a)RT—(1.162892x105(1.8.8801x10- a-9.6356x10'18

5

1.3xlo'1

 

5.58080474RT + 2.284467753x10
(20)

By appropriate manipulations, (20) becomes:

28

VOUT = 1.8235x10-1(1 + 2.785x10-18 - 3.27x10-la) x

P

. -1 -1 I
RT _ [1.48555x104 (1 + 8.88x10 -% - 9.636x10-18)
7 1 + 2.785x10 B - 3.27x10 a

RT + 4.093xlo4

(21)

 

 

L.
If a = 8 = O, the bridge output is zero when RT = 14.855k9. The
zero point can be shifted by appropriate adjustment of the G and I3 of
the potentiometers. The range of zero shift extends from 0.43k0 to

41.0k9.
Substituting the expression (3) into (21) and algebraically ma-

nipulating the expressions, (21) becomes:

 

 

VOUT = 1.8235x10"(l + 2.785x10-18 - 3.27xlo’1d) x
P 1
exp(3.821‘103/1.) _ 1.0895x106 (I + 8.88x10 _1 9.636x10-18)
l + 2.785x10 B - 3.27x10
<———e 3 6 a (22)
L exp(3.82x10 /T) + 3.00223x10 j

 

 

The above expression relates the bridge output voltage to the
variations in absolute temperature as measured on the Kelvin scale. It
can be simplified, since once G.and B are determined, the terms which
contain them are constants.

Let

A = (1 + 2.785x10-18 - 3.27x10—la)

l 1

B a (I + 8.88x10- cx- 9.63x10' 8
1

1 + 2.785x10-18 - 3.27x10" a
Equation (22) then simplifies to:

3 6
v = 1.8235xlO-1A exp(3.82x10 /T) - 1.0895x10 2

(23)
OUT exp(3.82x103/T) + 3.00223x10

a]

ha

115

Ill]
ba

39.

29

The previous expression can be solved for the absolute temperature

in terms of the bridge output voltage, the result being:

 

 

 

 

TOK = 2.561203x102 1 _2 (24)
_2 ‘26.6l7x10 )AB + VOUT
1 + (6.70472x10 )ln -1
L [51.8235x10 )A - VOUTJ

To express the above in degrees Celsius, subtract 273.150K from

the above.

T°c = T°K - 273.15°K

= 2.563x102 1 , _2 l -2.7315x102
(6.6174x10 )AB + vOUT

 

1 + (6.7047x10-2)In

 

(1.8235x10-1)A - v

 

 

OUT J

' (25)
Conversion to the Farenheit scale is accomplished by the following

formula.

T°F = %(TOC) + 32°F (26)

The temperature vs. voltage expressions developed above are used
when executing data reduction, negating the construction of any complex
hardware linearization circuitry. Since linearizing circuitry is not
used, power consumption is considerably reduced.

Construction of the bridges commences with determination of com-
ponent layout upon a 4%" x 6%" glass epoxy vector board. To fit the
boards within the aluminum casting, it was necessary to reduce their
dimensions to 4%" x 3%". They then plug into sockets mounted on alumi-
num stock 1/4" x 3/4" x 9". The sockets were then wired forming a
backplane facilitating interconnection changes. The vector board it-
self has 44 metallic contact fingers (22 per side) along one edge of

the 4%" width. These fingers provide the mating connections to the

30

socket terminals. The set of fingers on one side are labeled alphabeti—
cally, while those on the other side are labeled numerically.

Since a maximum of 18 temperature measurements are desired at any
one time, a corresponding equivalent number of bridges had to be con-
structed. Conservation of available space, as well as a savings in the
quantity of components, is realized through construction of two boards
each with nine bridge networks, all sharing the same reference leg.

Also included as part of the board circuitry is a voltage divider net-
work used to monitor the battery supply potential. Figure 7 diagrams
the circuit topology which is implemented using wirewrap connections.
The components are mounted on the alphabetically labeled side of the
vector board and wirewrapped together on the reverse side. The contact
finger terminations are listed in Table V. To implement ease of adjust-
ment, the potentiometers are mounted along the top of the vectorboard,
thus alleviating any need to remove the boards from their sockets.

Oscilloscope observations indicated the presence of power line
frequency interference. This interference was attenuated by the addi-
tion of SuF capacitors across the output terminals of the bridge.
Capacitors were also installed from the output terminals to the negative
terminal of the battery.

Chapter IV presents the measured response of the circuitry as well

as the field data recorded.

3.2.2 Humidity Measurement

 

The sensor providing relative humidity indication is manufactured
by the Physical-Chemical Research Corporation [5]. It is an electric

hygrometric circuit element which senses changes in relative humidity

UH96I

 

UH96I

 

61961

‘N

 

 

31

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

@a@a@t@.@a®t@s@a@t

 

 

 

 

23319671

 

Electrical Topology of Thermistor Board

Figure 7.

Tner

The:

Th6!

The:

The

The

The

Hot

For

Thermistor

l

Thermistor

2

Thermistor

3

Thermistor

A

Thermistor

5

Thermistor

6

Thermistor

7

Thermistor

8

Thermistor

9

Monitor

Power

GI

32

TABLE V(a)

Thermistor Board Terminations

Thermiscor Board #7

 

No Connection
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Power Supply Ground
Reference Leg Bridge
Ground

Power Supply Ground
No Connection

Power Supply Ground

10

11

12

13

14

15

16

17

18

19

20

21

No Connection
Output Bridge 1
Thermistor
Output Bridge 2
Thermistor
Output Bridge 3
Thermistor
Output Bridge 4
Thermistor
Output Bridge 5
Thermistor
Output Bridge 6
Thermistor
Output Bridge 7
Thermistor
Output Bridge 8
Thermistor
Output Bridge 9
Thermistor
Power Supply Monitor
No Connection

Power Supply +

The

The

The

Th.

Thermistor

l

Thermistor

2

Thermistor

3

Thermistor

4

Thermistor

5

Thermistor

6

Thermistor

7

Thermistor

8

Thermistor

9

Monitor

Power

MMMMMMMMM

33

TABLE V(b)

Thermistor Board Terminations

Thermistor Board #8

No Connection
Output Bridge 1
Power Supply Ground
Output Bridge 2
Power Supply Ground
Output Bridge 3
Power Suppply Ground
Output Bridge 4
Power Supply Ground
Output Bridge 5
Power Supply Ground
Output Bridge 6
Power Supply Ground
Output Bridge 7
Power Supply Ground
Output Bridge 8
Power Supply Ground
Output Bridge 9
Ground

Power Supply Ground
No Connection

Power Supply Ground

10

11

12

13

14

15

16

17

18

19

20

21

22

No Connection
Reference Leg Bridge
Thermistor

Reference Leg Bridge
Thermistor

Reference Leg Bridge
Thermistor

Reference Leg Bridge
Thermistor

Reference Leg Bridge
Thermistor

Reference Leg Bridge
Thermistor

Reference Leg Bridge

[Thermistor

Reference Leg Bridge
Thermistor

Reference Leg Bridge
Thermistor

Power Supply Monitor
No Connection

Power Supply '

34

through changes in impedance. A chemically treated styrene copolymer

is processed into a plastic wafer. This nonconducting substrate is then
integrated with an electrically conducting surface layer, and any
changes in relative humidity cause its surface resistivity to vary.

The process of vapor absorption is slightly temperature dependent.
Fortunately, the coefficient is approximately linear over a wide range
of temperatures and relative humidities. For a given sensor resistance,
it is necessary to add 0.22/OF (percent relative humidity per degree
Farenheit) to the value indicated on the standard curve when the mea-
surements are taken in temperatures above 77°F; and for those measure-
ments taken in temperatures below 77°F, subtract 0.22/OF from the
standard. The standard resistance versus relative humidity curve is
given in Figure 8.

When designing the sensor into any circuitry, the following maxi-
mum ratings must be considered:

1. Only ac voltages of at least 20 Hz with zero dc component

should be used (a sinusoidal waveform is preferred).

2. Maximum allowable current through the sensor is 1.0 mA.

Any dc component placed upon the sensor will generate polarization
effects which shift the sensor calibration.

The sensor resistance undergoes a change of over four orders of
magnitude as the relative humidity varies from zero to 100%. From an
extremely high resistance of 40 M9 to a low of 1.3 kW the relative
humidity spanned is from 2% to 100%.

In order to make any meaningful measurement over the entire range,
it is necessary to switch varying values of resistance into the refer-

ence leg of a bridge configuration. Since ac voltages must be applied

35

 

lOMdp
lMJr
E
E
0
2
H
'5' 1001(1b
E
t—<
(D
H
(I)
m
I!
101(1-
1K 5 I l 3 4 ‘ i 5 #‘

 

C

10 20 30 (,0 50 6 7O 80 90 100
RELATIVE HUMIDITY IN PERCENT

Figure 8. Plot of Sensor Resistance vs. Percent Relative Humidity for
PCRC—ll Humidity Sensor

[0

re;
10*.
sad
be

det
jut
res
PUI
wit
the
hi}

Elli

no
0D
ti
tu
ta
C0
to
to

Dr

36

to the sensor, a scheme for generating the required excitation was im—
plemented. The ac signal magnitude varies as a function of the sensor
resistance, but in order for the logger to accurately record the output
of the bridge, the signal must be a slowly varying dc response. The
requirement can be satisfied if the ac voltage is rectified and then
low-pass filtered. If the reference signal is processed in exactly the
same manner as the sensor signal, then a high degree of accuracy can

be obtained. The difference between the two resulting dc signals then
determines deviations from bridge balance. The output signal in con-
junction with a window comparator determines if the proper range is
realized. If the signal is within the comparator window, then its out-
put disables the range control circuitry. When the signal strays from
within the window, the comparator generates a signal which activates
the control circuitry. The circuitry then searches until, as deter-
mined by the comparator, the proper range is found. The operation of
the aggregate components is established in the block diagram of Figure
9.

Complementary Symmetry Metal Over Silicon (COS/MOS or CMOS) tech-
nology has created both analog and digital building blocks which
operate over large temperature spans, with extremely low power dissipa-
tion, and which remain insensitive to power supply ripple and tempera-
ture changes. The circuits are able to operate from power sources
ranging in value from a low of 3.0 volts to a high of +15.0 volts.
COS/MOS digital circuits exhibit high noise immunity due in large part
to their slower operating speeds and the low output impedance of the
complementary gates. Different manufacturers have expanded their

product lines of COS/MOS devices offering units with increased circuit

Mzrmwz HHDK N>UWUH=-HI ‘

uHoopHu ucosousmmoz hquHasm mo EmuwMHQ HUOHm .m oustm

 

 

 

 

 

 

 

 

37

   
    

 

 

 

 

     

 

     
  

 

 

 

 

 

 

 

 

 

 

 

 

SEE A
owsmm
on
m >+
moumoanH
owsmm
mwooouomsm,
Houmumoaoo
, Housemano Houusoo Houmfiwom
uMHnm
soust
monHuoom Macauoz
o>m3 HHsm Houmwmom
on .
m >I —
. HouHHm
\\\\\\\\ _mmmm 30H
Houmscsuu< ll/I/l/I moumHHHom

 

 

 

“3335. $03....
mmmm 3OH

 

 

 

 

uonHuoo. possum
0>m3 H H Pm Ufi—ufigm

 

C01

C3

of
ra
of
tr
C0

si

at
tr

na

sh

UP

51
Uh
9h
50
re
ne

CD

38

complexity. These various device types, their characteristics and typi-
cal application ideas, are completely documented in publications by RCA
[6],Motorola [7], and Fairchild [8].

Having pointed out many of the practical features of COS/MOS IC's
(integrated circuits), various devices have been incorporated into much
of the humidity sensor signal conditioning electronics. For example,
range indication is implemented using a resistor ladder network, COS/
MOS CD4016 Quad Transmission Gate chips and a power supply. A schematic
of the components' interrelationships is given in Figure 10. The con-
trol signals to the various gates are derived from a CD4017 Decade
Counter/Divider. The counter, when driven by a clock signal, succes-
sively presents a "high" signal at the various output pins labeled 0-9.
A "high" signal corresponds to the presence of the power supply value
at the output. This "high" logic signal energizes the corresponding
transmission gate which passes the ladder voltage to the output termi-
nal. The ladder voltage ranges in value from -50 mV to +50 mV in steps
of 10 mV. An output of -50 mV signifies that the "0" state of the
shift register is high and a one-to-one correspondence is established
up through state "9" when the output will be +50 mV.

The control signals which activate the range indicating transmis-
sion gates also affect corresponding gates which form part of the bridge
which contains the humidity sensor. An oscillator energizes the bridge
whose composition includes parallel combination of the sensor and a
500 k0 resistor in one leg, a 1.1 k0 resistor in series with a 200 k9
resistor in each of two other legs, and a resistor-transmission gate
network in the fourth and final leg as depicted in Figure 11. The same

CD4017 Decade Counter/Divider that controls the gates in the range

39

 

xsosooz wsHumoHvsH owcmm mo uHsouHo poHHmuon .OH omome

mouse COHmmHEmsmHH I as

aon GHOH GHOH GHOH GxOH axomH GZH

3m:

0 :w: 0 93$: 0

:5:

:Q: :m: o :N: s :H: a :0: o

as. .

 

  

w

 

H
O

 

 

GxOH GHOH GHOH GHOH GHOH axomH SSH

mwvHOOm. ‘ LN I 2%

I

4O

owmem oHumaou3¢ mo mHHmuon .HH ouswwm

 

 

 

 

Recess uaaunao Houuaou finauneeaaoo aneHHaae Haanameneean
suoHo aHedam suoHo _

 

 

 

 

 

   

emanauodm

 

ndaeeaodm
a>ns HHas

«Heads
xuoHo

U
sooHu seaweeds Sufism NHoeno summo HHoenox fiHHVAYIIIIII.

:02 Ill l E I

2m: 2w: ..N: :0: 2m... =¢V= :m: :N: :H:
H.H
\ L . [fl/ W W 93.4

0>N3 HHDm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

II - I» I) I I - II lrIi. eHoenu
+ I acumHHHomo
a=o> I ,
mos woe use was nos ace mos Nos Hue owe
.
v
mm mm am as me em mm N. as om .m.

axoom

 

ind
nor
the
she
the

dec

th

th

th

41

indicator also controls the gates in the resistor-transmission gate net—
work of the bridge. Transmission gates shunt the 200 RR resistors in
the bottom legs of the bridge to increase the sensitivity of the bridge
when the sensor and resistor network have low values of impedance. Then
the transmission gates are "on", shorting the 200 RR resistors, which
decreases the effective voltage between the bridge outputs and ground.
These gates are controlled by the "carry out" pin of the CD4017, and
when the lower-valued reference resistors are connected in the bridge,
the carry out is "low". It is inverted and applied to the controls of
the shunts, allowing them to short out the 200 k9 resistors when the
"carry out" is low.

The oscillator is comprised of a CMOS CD4047 Astable/Monostable
Multivibrator operated in the astable mode; a CD4016 Quad Transmission
Gate chip, which functions as a current switching device; and a CD4007
Dual Complementary Pair plus Invertor. The CD4007's main function is
that of power-up control. When the glass epoxy board, on which the
above active devices are mounted, has been inserted into the power
socket, the batteries are connected to the devices. With initial con-
nections, two rechargeable 6.25 900 mArhour batteries are connected in
series. The most negative potential connects to the -VSS terminations
of the CMOS devices, while the most positive potential terminates upon
+VDD pins of the CMOS. By referencing the CMOS outputs to the point of
connection between the two batteries, symmetrical positive and negative
going waveforms are generated. Initial power up procedure is hardware
designed to minimize device failure due to improper application of

signal sources. To avoid possible COS/MOS device destruction, it is

imperative that all signal sources be disconnected from them before

42

the power is removed from such circuitry. Similarly, the circuitry
should be powered before applying any signals. For the above reasons,
an oscillator hold off circuit was constructed which disables the os—
cillator until a capacitor reaches the switching point voltage value of
a COS/MOS CD4007 invertor. When the invertor output goes low, the
oscillator is enabled. The oscillator acts as the synchronizing ele-
ment of all Other COS/MOS circuitry. It functions as the clock input
for the CD4017 decade counter/divider (derived from pin 13), and the
outputs from pins 10 and 11 act as the gate controls for the CD4016
Quad Transmission gates. The outputs at pins 10 and 11 are guaranteed
to have a 50% duty cycle, and the output at pin 10 is shifted 1800 from
that at pin 11. The gates controlled by pin 10 are therefore on for
50% of the time, while the gates controlled by pin 11 are off; and for
the other half of the period, when the gates connected to pin 10 are
off, the gates connected to pin 11 are on.

The transmission gates form a switching network which alternately
connects the positive potential and then the negative potential of a
series combination of 1.5 V nickel-cadmium batteries. Via the above
method, an alternating frequency signal of 500 Hz and zero dc component
is derived to energize both the bridge network and the humidity-sensing
transducer. The power supply as constructed upon one glass epoxy vec-
tor board is depicted in Figure 12.

The voltage which is developed across the network leg consisting
of the parallel combination of the transmission gate and 200 KO resis-
tor in series with the 1.1 kg resistor, is compared with the voltage
developed across the similar network leg in the other branch of the
bridge. Before comparison, each voltage is precision rectified and

and low pass filtered, resulting in a pure dc signal.

PUU gwflVH fiUEAV

H-

>m~.o

zHaasm Hozom was HoumHHHomo

. NH 3&3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Homo HHomo

II\ _ .Iv
fillAAfl IA“. IWVVI I ./ Hfifiw VVYII
l + o

. . o . >m H .
>m H >2 6- HF >2 a+ >8 6-
man
- I \H I - SS 8 2H Jl
ENGIVl, a a HH 2 e a .
. a Hands 2
_.l H
naoom Heoe no
H
H H.H OH >m~.6+ e688
w a
>m~.e+
H J. .H HI
r _ _ P
HH w e H
2 r e l HHL
>m~.a-
v i. OH H H N 3.
r h a 8% “ 55+
H - .6 ,

cis

CU]

ini
op-
p05
Bet
iat
upo
inp
Sig

pro

as

Sum
ten
P08:

dtl

amp]

44

Several techniques exist which use various devices to perform pre-
cise rectification I9]. [10]. The configuration finalized in the cir-
cuit of Figure 13 is constructed using the following components:

2/3 Siliconex Micropower Triple Operational
Amplifier Integrated Circuit (device type
L144A),

2/6 RCA Ultra Fast Low Capacitance Matched
Diode Array (type CA3039),

5 196 k0 1% Precision Resistors, and
2 .l uF Ceramic Capacitors.

The circuit operation can be explained as follows. Making an
initial assumption that the diodes are nonconducting implies that the
op-amp is operating in the open—loop mode. Any input at the
positively—designated terminal of the op-amp is compared with ground.
Because of the high open-loop gain, the amplified output causes immed-
iate conduction of one of the diodes. Which diode conducts depends
upon the polarity of the input signal with respect to ground. If the
input signal is positive, diode (:) conducts and for a negative input
signal, diode conducts. Immediately upon conduction, the feedback
provided by the diodes forces op-amp into closed-loop operation.

In the case of conduction through diode <:) , op-amp l | operates

 

as a voltage follower and the output mimics the input. Op-amp [a
sums the voltages at its inputs. The voltage applied at the negative
terminal is amplified by a gain factor of -1, while the voltage at the
positive terminal is amplified by a gain factor of +2. The net result
at the output terminal is an exact replication of any positive input
signal.

When diode conducts, op-amp operates as a non—inverting

amplifier with a gain of 2. Op-amp [a again operates as a summing

45

uHsoHHo HoHMHuoom conHoon

 

.mH mostm

 

 

 

 

 

 

 

 

 

 

 

 

zH>
I Av
eeeHHm H
56> \ + $3.5 + I
6 E E

Q I @633

A

3628: @W _
anomaue\H
IIIl))>I.|II\</>l I3? {(111
asGAH seed aseaH asOAH

46

amplifier. In this case, however, the signal at the negative terminal
is amplified with a gain of -2 and that signal applied to the positive
terminal is amplified by a gain of +3. Since the signal applied to the
negative terminal of the summing amplifier has already been amplified

by op-amp , the actual output due to just the signal at the nega—

tive terminal is -4 VIN' where V

IN is the input voltage to the precision

rectifier. When added to the output due to the signal at the positive
terminal alone (that is, +3 VIN" the net output is (3 VIN + -4 VIN = -VIN).
Recall that the input signal was negative with respect to ground, so
that actual output result is positive. If the diagram of Figure 13 is
decomposed into the two cases of positive and negative voltage inputs,
as indicated in Figure 14(a) and 14(b), the analysis is easier to
follow. Next, a further subdivision is performed by introducing the
Thevenin equivalent voltage sources in Figure 14(c) for a positive
signal and the Thevenin equivalents in Figure 14(d) for negative signals.
USing superposition, the results indicated above are easily obtained.
When a negative signal is applied, op-amp is switched into
a configuration in which the output voltage is double that of the input
voltage. To prevent clipping of the output, it is important to restrict
the input signal to magnitudes less than one-half of the clipping level
voltage. The op-amps are powered by bipolar supplies of magnitude 6.25
volts, and since the bridge signal is derived from a bipolar supply of
1.3 volts, the rectifier operation is well within the bounds of the
above constraint. Two identical networks were utilized in order to
minimize signal shifts resulting from component value variations im-
parted by temperature changes. After rectification, the processed

signals are amplified by a differential amplifier which incorporates

HonHooom conHoon onu mo coHumuoao .QH oustm

MHmcme o>Humwos pom muamHm>Hsco cho>o:H A.m

HBO mHmame o>HuHmon Mom muaon>H=uo CHco>o£H A.o

 

 

axooH axoaH axooH

 

EDD

47
?:>

 

cowumuswwmcou HmawHw o>Humwoz A.n

 

chmH axomH axomH axoaH

 

:oHomustmaoo Hmcme o>HuHmom A.m

 

 

 

axomH axomH axomH axomH Boo

 

axomH axoaH

axomH

 

48

input filtering to remove the high-frequency hash generated during the
rectification process.

The precision rectifier output obtained from the resistor-
transmission gate network leg passes through a low-pass filter con-
structed of a l'kQ precision resistor and a 5 MP capacitor and the
signal is then applied to the positive terminal of the differential
amplifier. Similarly, the rectifier output obtained from the sensor
leg of the bridge passes through a low-pass filter and is applied to
the negative terminal of the differential amplifier. The gain of the
differential amplifier has been adjusted to a value of 10 by appropri-
ate choice of component value as depicted TIL Figure 15. The output
of the differential amplifier provides the input to the window compara-
tor. This input voltage is developed across a 20 k9 potentiometer load,
which serves the dual purpose of a voltage dividing attenuator and
filter. The 20 k0 load resistor is isolated by the voltage follower

(Z) in the window comparator circuit. The other two op-amps perform
the comparator function. The low level reference applied to op-amp <:)
sets 'its trim) point and the high level reference sets the trip point
for op-amp (:) . When an applied signal is within the comparator
window as defined by the reference voltage, the outputs of op-amps (:)
and (:> are both high. Since these outputs are directly connected
to the inputs of one of the four two-input nand gates on the COS/MOS
CD40ll IC, the output of the nand gate is low. The nand gate output
is connected to the clock input of the CD4017 Counter/Divider and the
clock signal is applied to the clock enable signal. By the stated
change of connections, the counter/divider will only increment if the

output of the comparator is "high;" that is, +V The nand gate

DD'

49

.xuoBHoc HouHHm ammo 30H com HoHMHHmEm HMHucoHoMMHQ .mH omstm

 

 

llgi

 

GMON

 

GMOOH

 

 

 

I»

.uw

 

\
/

 

 

 

 

GMOOH

GMOH GMH

 

(I

50

output goes "high" whenever the input signal leaves the window bound-
aries of the comparator. The comparator circuit was derived through
slight modifications of a similar circuit described in application
literature [11]. Both the upper and lower reference voltages are de-
rived from circuitry which utilizes two RCA CA3078AT Micro Power Op-amps.
The circuit diagram of Figure 16 lays out the component interconnections
of the reference sources. The circled numbers in the diagram indicate
the designations of the leads which protrude from TO-5 cans in which
the IC's are packaged. The op-amp marked (:> is connected as a voltage
follower. This amplifier takes the output from the potentiometer and
isolates it from the rest of the circuitry. The follower output pro-
vides the upper levellinfljzto the window comparator, and by cascading
with an inverting op-amp, the lower level reference is generated. The
above provides a positive reference and a negative reference of equal
magnitude. By varying a single potentiometer, the reference voltages
will track one another, allowing symmetrical variation of comparator
window width.

The circuitry comprising the transmission gates, resistor networks,
bridge components, precision rectifiers, and the difference amplifier
is constructed upon a glass epoxy vector card. Because of the lack of
any more useable space on the card, construction of a second card pro-
vided a foundation upon which to construct the voltage reference and
window comparator circuitry. Since the power supplies energize other
circuitry in addition to those which provide humidity indication, they
were constructed upon a separate epoxy vector board. The entire cir-

cuit is depicted in Figure 17.

51

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tn...
l96kf2 fO-MHF
329a ____ 19....
l \\¢<:) 196k0
®l \ © 196m
v. o
1N ®
10m @ (D v
l96kQ
VQD UL (3)
0.01% “mag
T-LSV :’ 0.01uF

(a) Reference

 

 

f
.04" F
I“ 1: mo V01.

 

 

 

 

Comparator

 

 

 

Output

 

 

From ©
Differential

C} +
Amplifier

. 047pF I

 

 

 

 

 

 

 

 

(b) Window Comparator

Figure 16. Reference Sources and Window Comparator.

52

uHsouHo ucoaouommoz mquHfism ouonEoo

u

no.

. NH 3:me

Lx

ea.

no.2.

A;>n«4+

 

53

As mentioned previously, the reference leg of the bridge consists
of 10 series combinations of resistors and transmission gates. The
values of these resistors were determined to a first approximation by
the following method. A simplified bridge structure was formulated
assuming that the transmission gates were ideal switches. Next, for the
lower values of sensor resistance, the 200 RR resistors were removed
from the bottom legs of the bridge. Using voltage division, the bridge
output, as a function of the resistances, was developed. In this
simplified case, the equations are similar to those developed for the
thermistor bridges. Continuing with the analysis, a constant dc supply
is assumed to energize the bridge whose magnitude equalstjmapositive
magnitude of the oscillator signal. Then it is assumed that ~10 mV is
the output of the bridge. The bridge equation is then solved for the
value of the reference leg resistor. Using the determined reference
leg resistor value and an output signal of +10 mV in the bridge equa-
tion, a new value of sensor resistance is obtained. Assuming that the
bridge output voltage is now ~10 mV, the sensor resistance at +10 mV is
used to obtain a new value for the reference leg resistance. The above
steps are repeated until all ten reference leg resistances are obtained.
The actual values in the constructed circuit were determined using the
method above, but at the laboratory level. A potentiometer represent-
ing the sensor resistance was measured and inserted into the circuit.

A reference potentiometer was inserted into the other leg of the bridge
and adjusced until the output of the differential amplifier exactly
equaled ~.l volt. The reference was then connected to another trans—
mission gate and the procedure repeated for increasing values of the

sensor potentiometer.

54

With the determination of the reference leg resistance values, the
humidity sensing circuitry design is completed. The reader will find
measured data concerning its operation in Chapter IV. This section con-
cludes the development of signal sources which are inputted to the log-

ger. The following section examines the characteristics of the logger

itself.

3.3 The Data Logger

 

Another important component of any data logging system is the re-
cording unit. The recorder preserves the data. A modular unit (Incre—
Logger; see Figure 18) establishes the intermediate function of data
record storage and a magnetic tape provides a convenient medium for
data storage. Compactness, ruggedness, self-contained power capability,

and large capacity recording are all important characteristics possessed

by the recording unit.

Taking the block diagram approach, as indicated in Figure 19, the
recorder can be described through a breakdown into nine subsystem func~
tions. Basic structural components consist of:

(1) 91923, The clock generates a continuous stream of voltage
pulses. The pulses, being equally spaced in time, supply an
accurate time reference. When appropriately counted, these
time pulses provide longer interval periods.

(2) Counter/Divider. The counter/divider processes the clock

 

pulses and provides an interval indicator. It determines
the time interval which must elapse before the next period
of data acquisition initiates. Also, it stores the elapsed

number of hours and days since the circuitry was last reset.

55

 

 

Exterior

 

Interior

Figure 18. Exterior and Interior View of Incre-Logger

 

 

 

 

 

56

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I

 

 

 

 

 

10 10
Analog Channel Analog
Signals Multiplexer To
Digital
Converter
10
Agalog Channel F
Signals Multiplexer
Digital
Analog
<é\\l//;§ignal
‘ 1 Recording
Head
Deck
Control
Logic Cassette
Advance
Scan Stepping
Switch ‘ InItIate1 Motor
Selectable
Divider Z 5
Board

 

 

 

 

 

 

 

Power

Supply

 

 

 

UJT

1 Clock

Figure 19.

Block Diagram of Incre-Logger

57

(3) Deck Logic/Power Supply. This circuitry provides both power
and control to the various subassemblies. Serial phase-encoded
digital signals generated by the deck logic integrate with
cassette motion to implement data storage.

(4) Multiplexer. The analog multiplexer sequentially selects each

 

data channel and presents it as input to the A/D converter.
Provision of a skip channel feature enables selected omission
of specific data channels.

(5) A/D Converter. The converter transforms analog signals into

 

corresponding binary representations. It implements the con—
version process via the single-ramp circuit configuration.

(6) Cassette Transport. The transport consists of a logic con~

trolled stepping motor which energizes a set of gears and
pulleys moving the capstan and tape take-up reel.

Random transistor transistor logic (TTL) performs all logic func-
tions necessary to the operation of the recording unit. Those circuits
which remain energized continuously are implemented using low-power
logic chips. Through the use of complementary unijunction transistor
and various low-power TTL logic integrated circuits, development of a
one pulse per second signal is realized. This signal is transferred
to the counter/divider board where it inputs to a user-definable count-
ing and dividing network. Every hour a four-bit binary counter is
incremented and, at the end of twenty-four hours, another four-bit
binary counter is also incremented. In this manner, elapsed time is
indicated. Concurrent with the counting, a switch selects a particular
divider output, routing it to a one-shot multivibrator. The resulting
signal closes a relay providing the trigger for the sequence of events,

resulting in data storage.

58

At the closure of the relay, the base of a power transistor receives
current; immediately upon saturation, activation of the remaining cir-
cuitry ensues. The following chart enumerates the user available trig-
ger intervals.

Switch Time
Position Interval

1 Sec.
5 Sec.
15 Sec.
30 Sec.
60 Sec.
5 Min.
10 Min.
15 Min.
30 Min.
1 Hr.

OKOCD'NIO‘Lfl-F-‘UJNI—

l'-‘

The main current board contains the power supply turn-on control,
an inverter, voltage regulating elements, and function synchronizing
logic.

A power transistor acts as a switch, turning the power both on and
off. Since many of the analog operations are performed by operational
amplifiers, they operate most efficiently when powered from a bipolar
supply of 15 volts, an inverter converts the unipolar 6~volt battery
supply to the higher bipolar voltage. A precision bipolar voltage
regulating network ensures stability of the op-amp's power supply. The
synchronizing logic, at turn-on, initiates itself, and the associated
data conversion functions. During conversion, the logic ensures proper
operation and timing of the various processes culminating in a correctly
encoded data word. Integrating the cassette drive mechanism with the
encoded data, the logic stores the information semi—permanently on
magnetic tape. A turn-off pulse from the logic initiates shutdown of

all circuitry exclusive of the clock, timing and counting.

59

The analog multiplexer sequentially directs each analog data chan-
nel to the analog-to-digital converter (ADC) inputs. The deck logic
generates an analog multiplexer enable signal, activating the multiplexer
clock. The clock inputs to a four-bit binary counter, whose outputs
drive a BCD-to-Decimal decoder. The decoded signals energize relays
which connect each data channel successively to the ADC. A last chan-
nel signal attains a high state indicating terminationwxfthe multiplexer
board function and relinquishing control of the deck logic.

The ADC in the recorder uses the single ramp-type conversion tech-
nique. In the conversion process, first an analog voltage signal in-
puts to a comparator and the converter digital-outputs reset to zero.
Secondly, a voltage ramp is generated providing a linear time increasing
reference for the comparator. As the ramp increases, clock impulses
input to a binary counter. When the ramp equals the analog input vol-
tage, the comparator changes state disabling the clock and preventing
further increase of the counter outputs. The counter outputs digitally
representing the input analog signal. For a large magnitude of analog
input, a correspondingly large binary number output rEalization results.
The slope of the voltage ramp, the clock frequency, and the maximum
count capability of the counter all play a role in determining the
maximum input voltage able to be accurately converted. The model cur-
rently used can convert voltages within the range ~51 mV to +51 mV.
(Voltages less than ~51 mV cause the converter to record zero input;
voltages in excess of +51 mV cause the converter to record the maximum
input of 377 ocral equivalent.) Bipolar inputs become realizable if
the input to the converter is offset by +51 mV. Thus, the converter

actually converts zero volts when the input is ~51 mV and 100 mV when

60

the input is +51 mV. After conversion, the binary number is bussed to
the deck logic. The logic processes the digital signal, eventually'
storing it on the cassette. A new signal transmits from the multiplexer
and the ADC resets so that the process may be repeated.

The cassette transport mechanism implements tape motion. To ensure
proper storage of the data, a stepping motor provides tape motion in
small precise increments. Pulses received by the stepping motor syn-
chronize with the data in order that data recording and tape motion
become concurrent events.

Several manual controls provide the user with convenience features
which assist one during operation of the equipment. A list of the fea—

tures follows:

 

Manual Control Function _
Header Thumbwheel Switches The Header Thumbwheel Switches allow
(Set of two; Channel One input of data by the user to label/
and Channel Two) date the set of data to be taken.

The data format consists of four bits
of BCD for each channel. Both chan~
nels are recorded as one 8~bit word
when this feature is used.

Enable/Reset Switch A return to the center switch which,
when held in the enable position,
allows the header data to be record-
ed when the write pushbutton is acti-
vated. When momentarily depressed to
the reset position, the counter and
clock circuits are all returned to
the initial state.

Write Pushbutton The Write Pushbutton initiates a
scan of all data channels, recording
them on the cassette whenever it is
activated. When the Enable/Reset
Switch is in the enable mode, this
switch, when activated, causes the
header data to be recorded.

61

In order to facilitate future cataloguing of logged data, cassettes
which are retrieved frdm experimental sites should be marked with identi-
fiers indicating test site location and the beginning date of the record-
ing. The header data feature of the "Incre-Logger" can be used to
identify recorded data. The 4~bit binary coded decimal (BCD) words can
be used to convey information directly when interpreted as computer dis-
play code. As an example: Set the header data thumbwheel switches to
"00" indication. Hold the "Enable/Reset Switch" in the header enable
position. Energize the "Write Pushbutton." Repeating the above pro-
cedure for the header message "TEST TAPE," the code relationships which
result are shown in Table VI. The string of five "OO"'s before and
after the Header words act as delimiters. The appendix provides charts
for cross-referencing the CDC display code with that of the header
words.

A specialized metal (aluminum) casting houses the entire electro-
nics and mechanics of the recorder, completely encapsulating the unit.
The casting consists of two parts: the top of the unit holds the re-
chargeable battery supply and the bottom is the recorder itself. An
O~ring seal has been imbedded in the cover to ensure an airtight closure
of the casting. A purge valve and vent screw allow dry nitrOgen to be
injected into the unit in order that damaging condensation may be pre-
vented. Blueprints of the logger circuitry and operating notes are

provided in the appendices.

3.4 The Reader System
As described previously, the recorder encodes the data and semi-
permanently stores that data on a cassette tape. To be useful, the

data must be retrieved from the tape and in some manner processed such

62

 

 

TABLE VI
Ilvnth2r lnItu I)IsI)lny lixnnu)lv
Header Word Qggal CDC Display Code

00 000

00 000

00 000

00 000

00 000

55 125 blank
24 044 T
05 005 E
23 043 S
24 044 T
55 125 blank
24 044 T
01 001 A
20 040 P
05 005 E
55_ 125 blank
00 000

00 000

00 000

00 000

00 000

63

that any useful information can be meaningfully presented to the re-
searcher. The recorder itself stores extensive quantities of data; as
such, any method of data retrieval employing a large amount of human
interaction quickly becomes exhausting. The tape reader system then
must perform the following functions to economically retrieve and pro-
cess data: (1) accurately decode the data stored upon the cassette,
(2) transfer the decoded data to a more permanent data storage medium,
(3) operate quickly upon large quantities of information, (4) provide
a method of data cataloguing and referencing while implementing ease
of data retrieval, and (5) supply the user with easy-to-use data ma-
nipulation algorithms (Statistical Analysis programs and plotting pro-
grams).

The heart of the reader system is the instrument which reads and
decodes data from the cassette magnetic tape. The "Incre-Data" Corp-
oration Model 1550 cassette reader transcribes, decodes and provides
computer compatible outputs of data recorded by "Incre~Data" Corporation
data loggers. The photograph in Figure 20 shows a front panel view of
the reader. Most of the control functions are positioned on the right
hand side of the faceplate, and data display options are situated on
the left side. Cassettes, which are to be read, are placed with the
cutout centered in the middle of the reader faceplate.

The control mechanisms and indicators located on the right-hand
side of the reader faceplate are identified as follows: Starting at
the top of the right-hand side, there exists a rotary switch allowing
the operator to determine transport "MODE" control of the tape move-
ment. The four functions available allow the operator to: (l)

"REWIND" the cassette; (2) read and transmit data from the cassette

64

 

 

 

 

Figure 20. Photograph of Cassette Reader

tape as it travels at a continuous speed of 12 inches per second across
the playback head, when in the "CONTINUOUS" position; (3) read at a
more relaxed rate of 3 inches per second (this speed allows the machine
time to generate control signals allowing precise tape positioning) when
placed in "INCREMENT" position; and (4) set the switch to "BACK ONE
SCAN," allowing the tape to be reversed and run backward at a speed of
3 inches per second. (NOTE: Data is not read in this mode.) Directly
underneath the transport "MODE" control lies a microswitch button
labeled "EJECT". This switch will eject a loaded cassette from the
reader provided the "MODE" switch is not in the "REWIND" position and
the "LOAD" switch is deactivated. The four multi-colored lights indi-
cate to the operator various reader operating conditions. The read
indicator signifies that the power switch immediately to its left has

been activated. The green light at the top of the column of indicators

65

notifies the operator that the cassette has been inserted properly into
the transport mechanism. The amber light directly below the green in-
dicates whether the tape is either at its beginning point prepared to
be processed, or that the end of the cassette tape has been encountered.
When the tape is in motion, data transmission is signalled by the flick-
ering of the blue indicator. The four rectangular pushbuttons light
up when activated. The uppermost button labeled "LOAD" engages the
playback head and pinch roller when the tape control is in any mode but
"REWIND". The "LOCAL" button must be energized to properly perform any
control functions from the front panel. When deactivated, the reader
can be controlled remotely via back panel cable sockets. When the
"ENABLE OUTPUT" switch is activated, data which is read transmits to an
output jack mounted on the back panel of the reader. Tape motion
commences or terminates when the "RUN" switch is momentarily activated.
When the tape is in motion, the switch light is on. Both speed and
direction of the tape is determined by the "MODE" switch position, in-
dependent of the "RUN" control. When the "MODE" switch is in the
"REWIND" position, activation of the "RUN" pushbutton has no effect.

The display options on the left-hand side of the reader facilitate
manual data recovery. Since large amounts of data are recovered, the
manual recovery can be a tedious chore. The display options do prove
useful when checking for data integrity or searching for particular
data. Use of these controls is described in the appendix under reader
operating instructions.

Once the data has been read from the tape, some scheme of trans-
ferring the data to computer storage must be worked out. The present

system employee a Texas Instruments "Silent 700" Data Terminal (see

66

 

Figure 21. Photograph of Data Terminal and Cassette Reader

Figure 21) equipped with an acoustic coupler. At the time of the
reader purchase, the necessary interface hardware was installed in the
Model 1550. The interface electronics assist in the data transfer and
permits computer control of the reader functions. In the transfer
process, a thermally written hard copy is produced by the T.I. "Silent
700" terminal as data is simultaneously transferred via the acoustic
coupler to the computer memory. An example of typical hard copy is
given in Figure 22, where the data is delineated as three-bit octal
words. The format for this data will be explained in the next section.
With the completion of the data transfer, the operator may optionally
select to resume control or terminate the computer link—up. If he opts

to resume control, the data terminal keyboard facilitates the trans-

mission of operator instructions. The data terminal also allows

67

000
000
000
000
040
010
011
031
125
011
031
125
001
125
040
005
O31
040
125
040
001
026
005
125
000
000
000
000
000
0000003770000000130310500661051231411572l4233251267306324342360377377
000000377000000014032050066105123141160215233251267306324343361377377
000000377000000013032050066105123142160215233251267306324343361377377
0000003770000000140320500671051231421602l5233251270306324343361377377
000000377000000014032050067105123142160215233251270306324343361377377
000000377000000014032050066105123142160215233251270306324343361377377
000000377000000014032050066105123142160215233251267306324343361377377
000000377000000014032050066105123142160215233251270306324343360377377
000000377000000014032050066105123141160215233251270306324343360377377
000000377000000013032050066105123141157215233251267306324343360377377
000000377000000014032050066105123141160215233251267306324343360377377
000000377000000014032050066105123142160215233251270306324343360377377
000000377000000014032050066105123142160215233251267307324343360377377
000000377000000014032050066105123142160215233251270306324343361377377
000000377000000014032050066105123142160215233251267306324343360377377
000000377000000014032050066105123142160215233251270306324343360377377
000000377000000013032050066105123142160215233251270306324343360377377
000000377000000014032050066105123142160215233251270306324343360377377
000000377000000014032050066105123142160215233251270306324343361377377.
000000377000000014032050066105123142160215233251270306324343361377377
00000037700000001403205006610512314216021523325l270306324343360377377
000000377000000014032050066105123142160215233251267306324343361377377
000000377000000013032050066105123142160215233251270306324343360377377
000000377000000013032050066105123142160215233251270306324343361377377

Figure 22. Typical Data Terminal Printout

68

on-line construction of programs by the operator. This feature allows
the operator to construct special data processing algorithms which he
applies to the previously stored information. In Section 3.5, some
typical data processing algorithms are presented. In addition, the
appendix to this paper includes detailed operating and maintenance in-

structions which apply to the aforementioned equipment.

3.5 Software Considerations

 

When developing data processing software, it is essential that
programs be created in a manner which promote flexibility. The system
must be capable of quickly responding to operative changes within the
system. Any software developed must maintain the basic information
flow designated by Figure 23. The system operator should be able to
quickly and comfortably interface with the software package. It should
provide a good description of the various manual controls and an out-
line of proper operating procedure. Provisions for automatically
reading and processing the cassette header data, if made, would facili-
tate maintenance of data records. The records would include initial
date and time of data collection, test site location and data channel
designations.

Any researcher is severely handicapped if he lacks the knowledge
to properly utilize analytic and scientific tools he may have access
to. Such handicaps can be overcome by software which supplies the
necessary information to the novice user. Mr. Richard K. Brandenburg,
with the Michigan State University Mathematics and Science Teaching
Center, has provided a simplified program "INTRO" which instructs

persons in the use of the Model 1550 Cassette Reader. A listing of

 

Attach Data
Log Program

 

 

    
     
    
  

  
   
    

Print
System
Documenta-
tion

 
   
 

Operator

amiliar wi .
- stem?

Insert
Cassette

  
   

 

 

  
    
   
 
 

 

   
   

Read

  

 
 

Print
Cassette
Header
Header
Data

   
      
   
 
     

Data

 

  

Read
Cassette
Data

 

Store

Catalog
Data

 

Figure 23.

69

 

Convert Binary
Data to Milli-
volt Equiva-
lents

 

 

 

Transform
Voltages To
Measured
Environmental

   

 

 

Parameters

   
    
   

Tabulate
Data

 

 

   
    

Processing?

Statistical
Analysis

 

 

Manipulat

User Data

 

ion

 

Data
Output

 

 

 

Plotting

Software Flowchart for Information Flow

70

this program and an example of its operation are provided in the tables
on the following pages.

The experienced user is capable of dispensing with the introduction
and proceeds directly to the data transfer process. The researcher
identifies each data tape processed by means of the header data feature
described in Section 3.3. A program which,utilizes the header data,
incorporates look-up tables to convey information regarding test site
location, specific measurement use, and data channel identification.

A typical header string, such as that given in Table VII, can be decoded
as follows:

The beginning string of 00's precedes all information. The first
header data word, other than 00, conveys the month in which the data
was recorded. The second and third header words give the day of the
month and the year. Two more header words of 00 separate the data from
the two header words which give the time when the recording started.
TWO more 00 delimiters follow and then two header words of 55 appear.
The consecutive 55's indicate that the next set of header words repre-
sent the user's name in display code format. Another consecutive
appearance of 55's signifies that the display code message has been
completed. The display code is the same code which is used in the
CDC 6500 computer for transmitting alphanumeric information. The dis-
play code delimiters are followed by two consecutive 00's. Next, two
header words give both the table and entry numbers from which addi-
tional alphanumeric data is obtained. The format is repeated until a
set of five consecutive 00's is encountered. This set of delimiters
signifies that all header data has terminated and further data is then

recorded by the various sensors.

Header Data
String;

 

00
00
00
00
00
02
15
76
00
00
15
35
00
00
55
55
02
17
25
03
10
05
22
55
55
00
00
01
01
00
00
02
01
00
00
03
02
00
00
00
00
00

up.)

WV

71

TABLE VII

Header Data Example

Information
Conveyed

 

Beginning
Delimiters

February
15th
1976

Delimiters

3:35 p.m.
Delimiters

Display Code
Delimiters

User Name in
Display Code

50313-103063

Display Code
Delimiters

Delimiters

Table 1
Entry 1

Delimiters

Table 2
Entry 1

Delimiters
Table 3

0".

Entry 2.

Ending
Delimiters

Octal
Equivalents

 

000
000
000
000
000
002
025
166
000
000
025
065
000
000
125
125
002
027
045
003
020
005
042
125
125
000
000
001
001
000
000
002
001
000
000
003
002
000
000
000
000
000

72

The tables and entries, mentioned in the header data description,
store alphanumeric data which might be too lengthy to place directly
on the tape. Information concerning test site location can be stored
in the first table. The second stores information to be appended to a
data analysis. The computer needs to be informed concerning the type
of data signal recorded on each channel. Another computer stored table
lists the channel designations and the computer uses this table when
implementing processing procedures. Table VIII gives examples of the
tables described.

After the header data has been processed, the channel data is trans-
formed from the octal representation stored within the computer into
first, the voltage signal originally inputted into the A/D converter,
and then into the parameter of interest. Those parameters are tempera-
ture, battery supply monitoring, relative humidity range and a relative
humidity signal.

Several routines have been developed to facilitate data reduction.
The packed BCD header data is transferred to the CDC6500 computer in a
three-digit octal format. Program OCTLHDR converts the three-digit oc-
tal number into its corresponding header word. Program OMVC allows the
user to convert the octal numbers representing the voltages processed
by the A/D converter into those voltages. In addition, since the vol-
tages themselves correspond to other physical parameters, the two pro-
grams TEMPLIN and HMDTY convert those voltages into their temperature
and relative humidity equivalents. Listings of these subroutines are
provided in the Appendix. The algorithms were created as subroutines

in order that they easily interface with any input/output programs.

73

o ousumuwdsos momwusm ohm o~.:u oH musomquth oomuusm oHHm o~.=u

 

 

m unaumuwaaoh oomwusm new mH.:o m musumudeoh momuusm ohm mH.:o
e ousumuoaamh oumuusm «Hm wH.:u w ouaumuoasoa oommuam mam wH.:o
vow: uoz Dz “H.2u n musumuodEmH oomuusm new “H.:o
com: uoz Dz 0H.:o com: uoz Dz oH.:o
mmcmx suHcassm u>HumHum «mm mH.:o new: “oz :2 mH.=u
Hmcme mquHsam 9>HumHmm mmm 4H.:u comm uoz Dz eH.=u
m ousumuma509 momuuam mam nH.=u can: uoz Dz as.=u
condo :H HHom ouaumuomsmw mmH NH.=0 o unaumuanmH commusm ohm NH.:Q
suave :H ousumHoz HHow mzm HH.mo canon :H HHow musumuwaaoh mmh HH.:u
m musumuodsme mommusm New oH.:u m ousumuoaaow oomwuam new oH.mo
H ouzumquEoH oummuaw Hem a .mo q unauouoaawe momwusm chm a .mu
uOuHcoz mHaasm zumuumm zmm m .mo n unaumuanoH momuusm mum m .:0
saw: s02 32 a .mo . h318: aHaasm zumuumm zmm a .:o
pom: uoz :2 c .20 Hmcme zquHesm m>HumHom mzm 0 .=0
pomp uoz :2 m .:o owcmm hquHasm o>Huanm xxx n .=0
comma :NH HHom musumuoasoe NmH q .:u N manuwuoaaoa oomwusm NHm q .xu
suave :e HHom ousuwuoasos Hwy m .:o suede :m HHom ouaumuoaaoe mmh n .:u
:uama :NH muauaHoz HHom NZm N .:u canon :o HHom masumuwasmh Hmw N .xo
nuaoa :o musumHoz HHom HZm H .20 H muaumumdsmh oommuam Hem H .:o

Ham Ham

monumqumoo Hmccmno
u oHpmH
uaom H .oH mvcoomm 00 .n vumnouo oHda< n.5uHam .uz .e
mouocHz on .m mvcooom om .q mcovumo Hmu:UH:oHuuo= mm: .m
mouscHz mH .w mocooom nH .m one honsaz vcom
mmuscHz n .n moGOUom m .N acoEowmcmz AUHHmsc hoods .N
muscHz H .o ucooom H .H ouHm vwom mcHHHou .H
Hm>uoucH cmom oEHH cowumoOH
m oHan .< oHpmH

mmHnme monoumnuousaaoo mo deaamNm

HHH> m4m<a

CHAPTER IV

CIRCUIT RESPONSE AND TEST RESULTS

This section presents curves which depict the thermistor bridge
output voltage as a function of temperature for both the ideal case
and the measured response of the actual circuit. In addition, data
which indicates the measured response of the humidity instrumentation
is provided. The output response of the Incre-Logger,reader and out-

put terminal is presented and evaluated.

4.1 Thermistor Circuit Response

 

The equation relating the thermistor bridge output voltage as a
function of thermistor temperature was developed in Chapter III.
Equation (22) expresses the relation, while at the same time incorpor-
ating the effects of various settings of the two potentiometers. Both
a and 8 represent fractions of the potentiometer settings. Figure 24
depicts the output voltage variation as a function of a, B and ther-
mistor temperature. The optimum setting indicated in the graph occurs
when a = B = 0.5. When a = .5, the potential measured between the po-
tentiometer wiper and the negative terminal of the battery is given as
100 mV. In the actual circuit the potentiometer was positioned so that
a potential of 91 mV was measured. The {3 of the other potentiometer
was set by inserting a 5.6 kg resistor in place of the thermistor and
adjusting the B potentiometer until a bridge output voltage of —43.6 mV

was obtained. Since the output voltage is also a function of the

74

OUTPUT VOLTAGE (mV)

75

 

5 j +
15 45
78.5 288.5 . 308.5 318.5

Temperature (°C and 0F)

a=0, B=.5

a .5
B: o B:

 

-60
Figure 24. Theoretical Thermister Bridge Output as a Function of Poten-
tiometer a and 8 Factors and Temperature.

76

battery supply voltage as depicted in Figure 25. The potential of the
supply battery varies as a function of temperature, electrical load and
the degree of charge held within the battery. As a result, it is neces-
sary to monitor the supply voltage in order to maintain a high degree

of accuracy. A voltage divider circuit consisting of one 196 k9 resis-
tor, one 10 k9 potentiometer, and one 10 HF electrolytic capacitor is
diagrammed in Figure 26. Eta represents the variability of the potentio-
meter. Using a battery with a mesured supply voltage of 1.364 volts,

the potentiometer was adjusted until an output monitor voltage of +43.6

mV was obtained. Since

 

 

n 10 k Vs
VOUT = 206 k (27)
I “ 206 k VOUT = (2.06 x 102)(4-36 X 10-2) (28)
l 10 k vs 1.364
n = .65847

If a 1.5 volt battery is used in the circuit, the output voltage
is 47.9 mV. By using the aforementioned value of eta, it is possible
to operate the bridge circuits from a variety of battery supply poten-
tials. It must be noted, however, that larger valued power sources
decrease the range of measurement of the bridge.

Having made the potentiometer adjustments, measurements of the
actual bridge response were performed. A procedure was devised which
uses a "Yellow Springs" proportional temperature controller, an accom-
panying thermistor probe, a combination heater and magnetic stirrer,
and 300 m1 of a 502 alcoholdwater solution to accurately record the
circuit response as a function of temperature. The 300 m1 of the

alcoholdwater solution was first cooled in the freezer compartment of

OUTPUT VOLTAGE (mV)

77

 

 

 

50«F
40-r
30'P
20.L
10 ul-
: : .L 5 4
~25 5 15 25 35 45
\ 78.15
248.15 258.15 268.15 x 288.15 298.15 308.75 318.15
-10-- Temperature (°C and 0F)
-2o.. °
4\
-30 ul- \
.\
e upply-1.0V
-40-- 9
.\ supp1y=1 0 1V
:upp1y=1.2V
-soi-

 

Figure 25. Graph of Theoretical Thermistor Output Variations with Supply
Voltage Variations

V
s

78

 

+
[—

1.3V

JJ
.1

 

Figure 26.

196kQ

 

(_
lOkfl jn I Supply

10 HF Monitor
I Voltage

Voltage Divider for Monitoring the Power Supply

79

an available refrigerator. Next, a laboratory thermometer was used to
measure the temperature of the solution. The solution was placed upon
the heater-magnetic stirrer and both the proportional controller probe
and the thermistor to be measured were placed in the solution. Using
the proportional controller to power the heater unit, while the mag-
netic stirrer maintained a uniform temperature throughout the solution,
the solution temperature was slowly raised. A Keithley 3%—digit multi-
meter continuously measured the bridge response, while at the same time,
the laboratory thermometer measured the solution temperature. Table

IX lists the recorded data points which were obtained using the above

apparatus and the graph of the data appears in Figure 27.

4.2 Humidity Sensor Circuit Responses

 

The calibration of the humidity sensor and the corresponding mea-
suring circuits was accomplished by the method described in Section 3.3.
A set of potentiometers were connected in series and substituted for
the humidity sensing element within the bridge circuit. Keithley multi-
meters monitored both the range and output signals of the humidity cir-
cuits as the potentiometers were varied over the span of 1.89 k9 to
3.59 Mg. This span corresponds to a relative humidity range of 10% to
892. The potentiometers were adjusted so that a 10 mV change occurred
at the output. They were then disconnected and their resultant resis-
tance was measured. Table X lists the various voltage-resistance re-
lationships and the equivalent relative humidity to which the resistance
corresponds. A plot of the tabulated data is given in Figure 28.

The recorder-reader system has certain input-output characteristics.

For any input voltage within the range -51.0 mV to +51.0 mV, the reader

80

TABLE IX

Bridge Output as a Function of Temperature

Voltage Out ij)

Tem

Voltage Out (mV)

 

 

m
e
T

840274947962848286248048246048264824047060

n1”
0999877665444332211000111222333445566777889
1 111111111111111111111111
+++++++++++++++++++..._....._._...._.......

46802468024680246802468246802468026804680468024
00011111222223333344444000011111222233334444555

111111111111111111111111
+++++++++++++++++++++++++++++++++++++++++++++++

8380846r38304626121741891..5026202694949351rJ93—1059
095518653332097765144432100001222233445566677888
43333222222221111111111

+++++++++++++++++++++++++ .___._......._.......

.UnzAUAU.Unu.4.U,o.4n2.u.4nu.4nzAU.8,o.4nznun2nun2.4,ono.Un4.4,o.8.U.2.4,o.8.unz.4,on8.unz.4,o
nunv.8.I.6.5.4.4.J.J.J.Jn2n2.l.l.1.UAunununu.u.5.5.5.).3,o,o,o,o,oqlql117:7.nonugunvnuolo,o,o,
_ . _ _ _ . . . _ . _ . . . . _ . . _ _ . + + + +.+ + + + + + + + + + + + + + + +.+ + + + +

Table IX (Cont'd)

0

Temp C

+ 9.8
+10.0
+16.0
+16.2
+16.4
+16.6
+16.8
+17.0
+17.2
+17.4
+17.6
+17.8
+18.0
+18.2
+18.4
+18.6
+18.8
+19.0
+19.2
+19.4
+19.6
+19.8
+20.0
+20.2
+20.4
+20.6
+20.8
+26.0
+26.2
+26.4
+26.6
+26.8
+27.0
+27.2
+27.4
+27.6
+27.8
+28.0
+28.2
+28.4
+28.6
+28.8
+29.0
+29.2
+29.4
+29.6
+29.8
+30.0

Voltagg Out (9y)

 

81

Temp C

+15.6
+15.8
+21.0
+21.2
+21.4
+21.6
+21.8
+22.0
+22.2
+22.4
+22.6
+22.8
+23.0
+23.2
+23.4
+23.6
+23.8
+24.0
+24.2
+24.4
+24.6
+24.8
+25.0
+25.2
+25.4
+25.6
+25.8
+31.0
+31.2
+31.4
+31.6
+31.8
+32.0
+32.2
+32.4
+32.6
+32.8
+33.0
+33.2
+33.4
+33.6
+33.8
+34.0
+34.2
+34.4
+34.6
+34.8
+35.0

Voltage Out (mV)

 

-l9.2
-19.4
-27.0
-27.2
—27.4
—27.8
-28.0
-28.2
-28.5
-28.7
-28.9
-29.0
-29.4
-29.6
-29.8
-30.0
-30.1
-30.3
-30.6
-30.8
-30.9
-31.0
-3l.4
-31.7
-3l.8
-31.9
-32.1
-37.2
-37.3
-37.4
-37.5
-37.7
-37.8
-38.0
-38.2
-38.4
-38.5
-38.6
-38.8
-38.9
-39.0
-39.1
-39.3
-39.5
~39.6
-39.8
-39.9
-40.0

Table IX (Cont'd)

0

Temp C

+30.2
+30.4
+30.6
+30.8
+36.0
+36.2
+36.4
+36.6
+36.8
+37.o
+37.2
+37.4
+37.6
+37.8
+38.0
+38.2
+38.4
+38.6
+38.8
+39.0
+39.2
+39.4

Voltage Out (mV)

 

-36.5
-36.6
-36.8
-37.0
-40.7
-40.8
-40.9
-41.1
-4l.2
-41.4
-41.5
-41.6
-41.7
-41.8
-41.9
—42.0
-42.1
-42.3
-42.4
-42.5
-42.6
-42.7

82

Temp C

+35.2
+35.4
+35.6
+35.8
+39.6
+39.8
+ao.o
+4o.2
+4o.4
+40.6
_+_z.o.8
:41.0
+42.0
+43.0
4.44.0
+45.0
+46.0
+47.o
+48.0
+49.o
+so.o
+51.o

Voltage Out (mV)

-40.2
-40.3
-40.4
-40.5
-42.8
-42.9
-43.0
-43.1
-43.2
-43.3
-43.4
-43.6
-44.1
~44.6
-45.2
~45.7
-46.2
-46.7
-47.1
-47.5
-47.9
-48.4

BRIDGE OUTPUT VOLTAGE (millivolts)

50+

40.

30-

20‘

10-

-10-

-20 II

-301

-404

-50-

 

 

Figure 27.

83

n l
I U

15 25 35

up

Temperature (0C)

Graph of Measured Thermister Response

1':
45

84

TABLE X

Resistance, Bridge Output Voltage, Range
and Relative Humidity Tabulations

 

Range V

Indicator OUT Resistance _B;§;
—50.8 mV 1.89 k0 89.0%
-40.5 mV 3.62 k9 75.0%
-30.0 mV 5.30 k9 67.5%
-20.0 mV 7.02 k9 63.5%
-10.2 mV 8.71 k9 60.5%

+50.9 mV - 0.7 mV 10.36 k9 58.0%
+10.1 mV 12.32 k9 55.8%
+20.2 mV 14.17 k9 53.9%
+30.0 mV 16.02 k9 52.25%
+40.l mV 17.93 k9 51.0%
+50.6 mV 20.00 k9 49.5%
-50.6 mV 19.6 k9 50.0%
-40.6 mV 21.5 k9 49.0%
-30.5 mV 23.5 k9 47.8%
-20.2 mV 25.7 k9 46.4%
—10.2 mV 27.7 k9 45.9%

+40.8 mV 0.0 mV 30.0 kg 45.0%
+10.4 mV 32.3 k9 44.25%
+20.4 mV 34.6 k9 43.8%
+30.0 mV 37.0 kg 43.0%
+40.0 mV 39.3 k9 42.5%
+50.2 mV 41.7 k9 42.0%
-50.5 mV 39.3 k9 42.5%
-40.4 mV 41.8 kfl 42.0%
-30.4 mV 44.4 k9 41.5%
-20.2 mV 47.0 k9 41.0%
-10.2 mV 49.5 k9 40.1%

+30.5 mV 0.0 mV 52.2 k9 39.5%
+10.3 mV 55.0 k9 39.1%
+20.1 mV 57.8 k9 38.6%
+30.5 mV 60.9 k9 38.2%
+40.4 mV 64.0 k9 38.0%
+50.6 mV 67.0 k9 37.5%
-50.0 mV 65.2 k9 37.6%
-40.0 mV 68.4 k0 36.8%
-30.0 mV 71.5 k9 36.5%
-20.3 mV 74.7 k9 36.2%
-10.3 mV 78.0 k9

+20.3 mV - 0.2 mV 81.5 k9
+10.l mV 85.1 kg

+20.4 mV 89.0 k9

85

Table X (Cont'd)

 

Ingizgior VOUT Resistance R.H.
+30.3 mV 92.7 k9
+40.0 mV 96.5 k9
+50.0 mV 101.6 k9
-50.2 mV 100.7 k0
-40.5 mV 104.6 kg 34.0%
-30.4 mV 108.7 k9
-20.1 mV 113.2 k9
-10.4 mV 117.6 k9
+10.1 mV 0.2 mV 122.4 k9
+10.3 mV 127.4 k9
+20.2 mV 132.4 k9
+30.0 mV 137.6 k9 32.0%
+40.5 mV 143.1 k9
+50.2 mV 148.6 k9
-50.2 mV 148.6 k9
-40.0 mV 153.8 k9
-30.4 mV 159.9 kn
-20.5 mV 166.2 k0
-10.5 mV 172.4 k9
-10.1 mV 0.0 mV 179.4 k9 30.0%
+10.1 mV 186.7 k9
+20.0 mV 193.9 k9
+30.3 mV 202.0 k9
+40.1 mV 210.0 k9
+50.0 mV 218.0 k9
-50.5 mV 215.0 k9
-40.1 mV ’ 225.0 k9 28.0%
-30.2 mV 233.0 k9
-20.2 mV 243.0 k9
-10.4 mV 252.0 k9
-20.3 mV 0.0 mV 263.0 k9
+10.0 mV 274.0 k9
+20.0 mV 285.0 k9
+30.1 mV 298.0 k9
+40.1 mV 311.0 k9
+50.2 mV 324.0 k9
-50.2 mV 327.0 k9
-40.2 mV 342.0 k9 25.0%
~30.0 mV 357.0 k9
~20.2 mV 373.0 k0
-10.0 mV 390.0 k0
~30.5 mV 0.0 mV 409.0 k9 24.0%
+10.0 mV 429.0 k9
+20.1 mV 450.0 k9
+30.1 mV 474.0 k9

86

Table X (Cont'd)

Range V

 

 

Indicator OUT Resistance R.H.
+40.0 mV 499.0 k9
’+49.9 mV 525.0 k9 22.0%
-50.1 mV 533.0 k9
-40.2 mV 562.0 k9
-30.2 mV 594.0 k9
-20.2 mV 630.0 k9
-10.0 mV 669.0 k9
-40.6 mV 0.0 mV 712.0 k9 20.0%
+10.1 mV 759.0 k9
+20.0 mV 812.0 k9
+30.0 mV 871.0 k9
+40.0 mV 938.0 k9
+50.0 mV 1.014 M9 18.0%
-50.1 mV 1.01 M9 18.0%
-40.1 mV 1.094 M9 18.0%
-30.0 mV 1.193 M9
-20.0 mV 1.31 M9
-10.1 mV 1.444 M9 16.0%
-50.8mV 0.0 mV 1.624 M9 15.0%
+10.1 mV 1.829 M9 14.0%
+20.0 my 2.09 M9
+30.1 mV 2.44 M9 12.0%
+40.0 mV 2.89 M9
+50.0*mV 3.59 M9 10.0%

30 «P

20 «-

 

-10 ,,

-204.

-3oa-

-40 u-IL

 

‘5” «r-

Figure 28.

87

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Measured Humidity Sensor Circuit Response.
indicate auto-ranges.

The RI's

88

output is a corresponding three-digit octal number representative of

the input voltage. The ideal relationship can be expressed as follows.

4.3 Data Logger Calibration
Given the octal number XYZ as the output of the reader, the input

voltage measured by the recorder is found.

-4 -2
(X-64 + Y-8 + Z 1) 4x10 5.1x10 VINPUT in volts (29)

Table XI summarizes the Incre-Logger output obtained when the indicated
voltages were applied to the various channels. The voltages at each
channel were measured with a Keithley model 168 digital multimeter. The
voltages were developed by a resistive ladder network powered by a 1.232
V nickel-cadmium battery. The theoretical and experimental results are
summarized in Table XII. 'The deviation is greatest for input voltages
which are negative.

After verifying that all components were operating properly, the
sensors, signal conditioning unit and recorder were transported to the
water quality management project site located south of the campus. The
recorder and signal conditioner was placed within a shed and the various

sensors were placed around its perimeter.

89

TABLE XI

Test Data Recorded on Digital Cassette Tape
by Incre Data "Incre-Logger" Model 5822 X/N 123

000001377015030044057072106121134147163212225240253266302315330344357
000001377015030044057072106121134147163212225240253266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121l34147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344357
0000013770150300440570721061211341471632l2225240533266302315330344356
000001377015030044057072106121134147163212225240533266302315330344357
000001377015030044057072106121134147163212225240533266302315330344356
00000137701503004405707210612l134147163212225240533266302315330344357

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m a

0 fit: mac:

a m m a a o m m

m a w m o m o u m o h

H a) mu w>uummauuu

m w u m o .2 m > .4 H u u u m .2 o a

tm°£8355$383°28255H§°

:2 <5 I3 P! hi #4 a: a: n: 2: E4 lfl lg &« h. h. a: a: n: 2: I3

O’HHHHHHHHHHHHHHH I-l

ad m w o w o a m o m m m m m m m '3 o '3 '3 '3

m m a a a a a a a a a a a a a a a a c a a :

85588888888888888888888
w o o .c .c .c .c .2 .:
88588888885888888888885

90

TABLE XII

Comparison of the Input Voltage
to the Recorded Voltage of the Incre-Logger

 

 

 

 

Channel Input Octal - Converted
Number Voltage (mV) Representation Octal ng17 Error (mV)
1 -44.8 015 -45.8 +1.00
2 -40.35 030 -41.4 +1.05
3 -35.9 044 ~36.6 +0.70
4 -3l.4 057 -32.2 +0.80
5 ~26.9 072 -27.8 +0.90
6 -22.45 106 -23.0 +0.55
7 -18.0 121 -18.6 +0.60
8 -13.45 134 -14.2 +0.75
9 - 9.0 147 - 9.8 +0.80
10 - 4.6 163 - 5.0 +0.40
11 + 4.4 212 + 4.2 +0.20
12 + 8.8 225 + 8.6 +0.20
13 +13.3 237 +12.6 +0.70
14 +17.7 253 +17.4 +0.30
15 +22.25 266 +21.8 +0.45
16 +26.7 302 +26.6 +0.10
l7 +31.2 315 +31.0 +0.20
18 +35.65 330 +35.4 +0.25
19 '+40.l 344 +40.2 +0.10
20 +44.65 357 +44.6 +0.05

CHAPTER V

CONCLUSIONS AND RECOMMENDATIONS

The purpose of this research effort was to develop, analyze, and
evaluate a system which can acquire data automatically at remote field
sites. The data is acquired in a computer compatible format which en-
ables the user of the system to quickly process the data. The equip-
ment used is relatively maintenance free for extended periods of time
and remains flexible enough to be easily reconfigured for many different
types of applications.

The current syStem configuration consists of a continuously powered
signal conditioning unit capable of operating for a period of two weeks.
Various sensors which provide the input signals to the conditioning
unit are processed and bused to a logger unit. The logger unit periodi-
cally samples the data, encodes it and stores it upon a magnetic cas-
sette. A 6.25 volt, 7 amp-hour battery enables the logger to operate
for a period of one week. The logging function is complemented by a
cassette reader which decodes the stored data, transferring it to mass
storage within the large computing unit on the M.S.U. campus.

The system, as a whole, is easily enlarged upon, as other trans-
ducers can be easily interfaced with the original equipment. The log-
gers are interchangeable and the signal conditioning unit is readily
reconfigured.

The temperature bridge networks, when used in conjunction with

the linearizing algorithms, are accurate and long lived. They are,

91

92

however, prone to noise disturbances which are picked up within the
high impedance components.

The humidity bridge network is especially unique in the approach
used to implement the measurement. Very accurate readings can be ob-
tained for relative humidity measurements above 60%. The present system
does suffer from rather low range definition at low percentage relative
humidities. It might be said that the sensitivity is too great for
percentages below 50%.

At the present time, the weakest link in the syscem is the logger
itself. It suffers two main deficiencies. One, the logger, while in
the stand-by mode, consumes far too much power in the time keeping
operations; and two, below freezing the power control circuitry mal-
functions, causing complete dissipation of all available battery power.

By replacing the clock and divider circuits by COS/MOS devices,
it would be possible to drastically reduce the power drain. The origi—
nal clock and divider circuits consumed a continuous 50 mA in the
stand-by mode. The clock itself uses 5 low power TTL IC's as well as
several discrete components, while the divider board is a continuously
powered behemoth of 12 lowbpower TTL IC's.

One of the loggers has had its clock circuitry replaced by five
COS/MOS devices. The COS/MOS clock was constructed from the following:

1 CD4047 Astable/Monostable Multivibrator

2 CD4017 Decade Counter/Dividers

l CD4013 Dual Typed Flip Flops

l CD4011 Quad Two Input NAND Gates

The components were connected as shown in Figure 29. The circuit

substitution has resulted in a measurable reduction in current drain

xooHo uoaoalaoH mOZU .mm mustm

vmh

n: H NWTN filv

ca H.Hoqnox.

.I._ .1.

soOHaNHSH-

 

 

 

 

7 i4“

 

 

93

 

 

 

Caz:
xQOHo mHos no H 888m
m. N. «H N umm
_.l_ .8
HO HO 5.
I +

 

 

oommm L j momma l—Hl momma
m 2 . . -

 

 

 

 

 

 

 

 

 

 

 

 

 

ma. w mH mH, NH m a m_
m

 

 

 

 

, m N H
o «(.4 —__.=vlu— >
> > M

9+ o+ 93$ mmomm

94

of 8 mA. If a corresponding COS/MOS divider board can be substituted
for the present low power TTL circuits, an even greater savings in power
can be realized.

The entire system as originally conceived was to operate for a
period exceeding several months. However, during the course of my work
with the system, I have come to feel that it is impossible to use the
current power sources to maintain circuit operation over the desired
time period unless some form of time multiplexing is incorporated.
Using COS/MOS transmission gates, it is possible to de-energize many of
the aforementioned circuits until just prior to sampling. After the
data has been logged, the circuits should again be de-energized. This
approach allows smaller resistances to be used for the bridge networks
which can decrease the amount of induced noise within the system.

Since the initialization of this project, many medium scale COS/
MOS IC's have been introduced. These IC's can greatly enhance the
potential of the system. One in particular is the 1-to-8 or 8-to-1
Multiplexer/Demultiplexer. The device lends itself to the construction
of simplified bridge switching networks such as that used in the
humidity-measuring circuit described previously.

In conclusion, the circuitry developed provides a stepping stone
to more accurate and longer liVed instrumentation. During the present
time period, it functions to provide scientists with data from remote
environments. From both a meteorological and agricultural point view,
the system provides a much needed service. When equipped with the
appropriate transducers, the unit has many uses in all types of remote
applications. The unit could conceivably be used to monitor the pol-

lution of lakes and streams. When discreetly placed in public places,

95

the unit could be used to monitor noise or air pollution. Indeed, the

uses are limited only by the imagination of the user.

APPENDI CES

APPENDIX A

LISTING OF SUBROUTINE "INTRO"

APPENDIX B

EXAMPLE ILLUSTRATING THE EXECUTION OF
SUBROUTINE "INTRO"

INTRO

THIS IS THE FIRST OF TWO FORTRAN PROGRAMS

DESIGNED TO AID IN THE READING AND INTERPRETATION OF DATA
RECORDED WITH AN INCRELOGGER/TRANSDUCER SYSTEM USING THE
1550 CASSETTE READER, PROCESSED DATA MAY BE SUMMARIZED

IN A TABLE,PLOTTED,AND/OR ADDED TO A NEW OR ALREADY
EXISTING PERMANENT OR MAG.TAPE FILE. SOME STATISTICAL
MANIPULATIONS OF THE DATA ARE ALSO AVAILABLE.

THE FOLLOWING INSTRUCTIONS SHOULD BE FOLLOWED
CAREFULLY IN ORDER TO READY THE READER FOR DATA
TRANSMISSION.

THE CONTROLS/SWITCHES SHOULD BE:

1. POWER SWITCH-ON. THE READ INDICATOR LAMP ALONG
WITH THE DISPLAY LAMPS SHOULD NOW BE LIT.

2. THE"LOCAL" BUTTON SHOULD NOW BE LIT.
IF IT IS NOT,THEN PRESS IT.

3. THE "LOAD","ENABLE OUTPUT",AND "RUN" BUTTONS
SHOULD NOT BE LIT. PRESS THE ONES THAT ARE LIT.

THE READER SWITCHES SHOULD BE POSITIONED

AS FOLLOWS :

SWITCH POSITION
TAPE/MEMORY TAPE(UP)
LATCHED/NORMAL NORMAL(DOWN)
INTERNAL/EXTERNAL/MANUAL EXTERNAL(CENTER)
WORD SELECT (NOT OPERATIVE)
REPEAT MEMORY OFF
DATA CONVERSION OCTAL
MODE INCRE

98

99

6. PRESS THE RESET BUTTON.

7. PRESS THE LOAD BUTTON ON.
THE TAPE WILL AUTOMATICALLY REWIND AND THE BOT
(BEGINNING OF TAPE) LAMP WILL LIGHT.

8. PRESS THE RUN BUTTON ON.

THE TAPE TRANSPORT WILL NOW BEGIN WINDING

THE TAPE ON THE LEFT REEL UNTIL THE BEGINNING OF THE DATE IS EN
COUNTERED. THE TRANSPORT WILL THEN STOP

AND THE BOT LAMP WILL GO OFF.

THE READER IS NOW READY TO TRANSMIT DATA.
PLEASE FOLLOW THE NEXT INSTRUCTIONS CAREFULLY:

WHEN THE TELETYPE PRINTS;

"END INTRO"
TYPE THE FOLLOWING RESPONSES,
ONE PER LINE;

1.0K-"SCRATCH"
2.0K-"ATTACH,A,INCRELOG."
3.0K-"USE,A"
4.0K-"RTL,IOO"
5.0K-"LENGTH,80."
6.0K-"RFL,IOOOO."
7.0K-"READPT,XXX."

YOU SHOULD THEN GET THE RESPONSE;
"READY FOR TAPE"

PRESS THE "ENABLE OUTPUT" BUTTON, SO THAT IT LIGHTS. THE DATA WILL NOW BE
TRANSFERRED T0 TEMPORARY FILE XXX.

WHEN DATA TRANSMISSION STOPS,THE BLUE DATA LIGHT
WILL NOLONGER BLINK, PRESS THE "ENABLE OUTPUT" AND
"RUN" BUTTONS SO THEY ARE NOLONGER LIT. THEN PRESS THE
"ESC" BUTTON ON THE TELETYPE. THE FOLLOWING RESPONSE WILL
BE PRINTED.

EDT-PROCESSING TAPE

OK-

TYPE THE FOLLOWING RESPONSES, ONE PER LINE;
1.0K-"REWIND,XXX."
2.0K-"PROMPT"
3.0K-"FTN."
4.0K-"LGO,XXX,YYY,INPUT,OUTPUT."

PROGRAM INCRELOG WILL THEN TAKE CONTROL.

END INTRO

APPENDIX B

EXAMPLE ILLUSTRATING THE EXECUTION OF SUBROUTINE "INTRO"

APPENDIX C

PROGRAM OCTLHDR

APPENDIX C

PROGRAM OCTLHDR

THIS PROGRAM CONVERTS THE OCTAL NUMBER REPRESENTATION OF THE HEADER
DATA INTO THE PACKED BCD FORMAT ORIGINALLY ENCODED
SUBROUTINE OCTLHDR(OCTL,HEADR)
DIMENSION IDGT(2)
THE BIT MASKS ARE GENERATED
JMSK1=MASK(4)
JMSK2=MASK(1)
IHEADR=O
ITEMP=OCTL
I=1
FOLLOWING STRIPS OFF 4 BITS OF EACH DIGIT
l IDGT(I)=ITEMP.AND.JMSK1
IF(I.EW. 2)GO TO 2
ITEMP=SHIFT(ITEMP,-4)
I=I+1
GO TO 1
2 DO 4,I=l,2
ITEMP1=0
ITEMP=IDGT(I)
DO 3,J=1,4
ITMP1=ITEMP.AND.JMSK2
lTEMP1=ITEMP1+ITMP1*2**(J-l)
3 CONTINUE
IDGT(I)+ITEMP1
4 CONTINUE
IHEADR=10*IDGT(2)+IDGT(1)
RETURN
END

100

APPENDIX D

PROGRAM OMVC

APPENDIX D

PROGRAM OMVC

TAKES AS INPUT 3-DIGIT OCTAL NUMBER AND CONVERTS IT INTO A
CORRESPONDING MILLIVOLT OUTPUT. RETURNS BOTH THE OCTAL NUMBER AND
THE CONVERTED MILLIVOLT EQUIVALENT.
THE CONVERSION xxx . SUM*4.0E-4 - 5.1E-2 IS APPLICABLE ONLY TO
INCRELOGGERS WITH BIPOLAR :51 MV INPUT RANGES.
SUBROUTINE OMVC(LLL,XXX)
SUM=0.0
XLLL=FLOAT(LLL)
1001 DO 1002,NNN=1,3
XLLL=XLLL/10
YLLL=AINT(XLLL)
SUM=SUM+((XLLL-YLLL)*10)*8**(NNN-l)
1002 XLLL=YLLL
xxx=SUM*4.0E-4+(-5.1E-2)
RETURN
END

0000

101

APPENDIX E

PROGRAM TEMPLIN

CO

00000

C500

APPENDIX E

PROGRAM TEMPLIN

THIS PROGRAM LINEARIZES THE TEMPERATURE-VOLTAGE RELATIONSHIP OF THE
THERMISTOR BRIDGE NETWORKS
SUBROUTINE TEMPLIN(TEMP11, REFl, TEMPER, MODEl)
TEMPII Is THE MILLIVOLT OUTPUT OF THE TEMPERATURE RECORDING CHANNEL
REFl Is THE REFERENCE MILLIVOLT OUTPUT WHICH INDICATES THE STATUS
OF THE BRIDGE POWER SUPPLY.
TEMPER IS THE CALCULATED TEMPERATURE, MODEl INDICATES WHETHER THE
TEMPERATURE IS IN DEGREES CENTIGRADE OR FARENHEIT
COMMON COEFFICS(ALPHA,BETA,GAMMA,ETA)
VSUP=(20.6/ETA)*REF1
VSUP IS THE SUPPLY VOLTAGE
ETA IS CALCULATED BY MEASURING THE SUPPLY VOLTAGE AND ADJUSTING THE
REFERENCE POTENTIOMETER so THAT THE REFERENCE IS FIFTY MILLIVOLTS
GAMMA1=TEMP11/VSUP
ALEPH1=(ALPHA1+1.2)/21.8+GAMMAI
RTEMP=(BETA1*(5.075E4)+3.75E3-1.0492E6*ALEPH1)/(e.56E1*ALEPHl-
(BET $A1+5.0))
RO=5.0E3
TO=2.9815E2
BETAO=3.873E3
TEMPPP=1/(l/TO+(1/BETAO)*LN(RTEMP/RO))
TEMPPP IS THE MEASURED TEMPERATURE IN DEGREES RELVIN
TEMPC=TEMPPP-2.7315E2
TEMPF=1.8*TEMPC+3.2E1
IF(MODEI.EQ.T)TEMPP=TEMPC
TEMPER=TEMPF
RETURN
TEMPC IS THE TEMPERATURE IN DEGREES CELSIUS
TEMPF IS THE TEMPERATURE IN DEGREES FARENHEIT
END

102

APPENDIX F

PROGRAM HMDTY

 

0000

('30

APPENDIX F

PROGRAM HMDTY

THIS SUBROUTINE CONVERTS THE RANGE INDICATING VOLTAGE AND HUMIDITY
SIGNAL VOLTAGE TO THE CORRESPONDING HUMIDITY MEASUREMENT. IT DETER-
MINES THE RELATIVE HUMIDITY THROUGH A PIECEWISE LINEAR APPROXIMATION
ALGORITHM

SUBROUTINE HMDTY(OCTLR,OCTLH,HUMDTY)
RANG=0.0
HVOLT=0.0

THIS REOUTINE USES SUBROUTINE OMVC TO CONVERT THE OCTAL NUMBERS TO
CORRESPONDING VOLTAGE EQUIVALENTS

CALL OMVC(OCTLR,RANG)

OCTLR IS THE OCTAL QUANTITY REPRESENTING THE RANGE
RANG IS THE CONVERTED RANGE INDICATING VARIABLE

CALL OMVC(OCTLH,HVOLT)

OCTLH IS THE OCTAL QUANTITY REPRESENTING THE BRIDGE OUTPUT

100
110

120
130

140

200

300

400

500

600

700
710

720

800

IRANG=INT(IOO*(RNAG+S.1E-2)+1)

GO TO(100,200,300,400,500,600,700,800,900,999)IRANG
IF (HVOLT) 120,110,110

HUMDTY= -(5.0/5.0E-2)*HVOLT+1.5E1

RETURN

IF(HVOLT+1.01E-2)140,130,130

HUMDTY= -(1.0/1.01E-2)*(HVOLT+1.01E-2)+1.6El

RETURN

HUMDTY= -(2.0/3.0E-2)*(HVOLT+4.01E-2)+1.8E1
RETURN

HUMDTY= -(2.0/5.0E-2)*HVOLT+2.0E1

RETURN

HUMDTY= -(3.0/9.01E-2)*(HVOLT+4.0lE-2)+2.5E1
RETURN

HUMDTY= -(2.0/7.02E-2)*(HVOLT+4.0lE-2)+2.8El
RETURN

HUMDTY= -(2.8/1.02E-1)*HVOLT+3.0E1

RETURN

HUMDTY= -(2.0/7.05E-2)*(HVOLT+4.0SE-2)+3.4E1
RETURN

IF(HVOLT+2.0E-4)720,710,710

HUMDTY= -(1.8/5.02E-2)*(HVOLT+2.0E-4)+3.6El
RETURN

HUMDTY= -(1.6/4.8E-2)*(HVOLT+5.0E-2)+3.76El
RETURN

IF(HVOLT+1.02E-2)320,810,810

103

810

820
830

840

900
910

920
930

940

999
960

950
980

970
991

990
992

993
994

995

104

HUMDTY= -(2.6/6.08E-2)*(HVOLT+1.02E-2)+4.0lEl
RETURN

IF(HVOLT+3.04E-2)840,830,830

HUMDTY= -(1.4/2.02E-2)*(HVOLT+3.04E—2)+4.15El
RETURN

HUMDTY= -(1.0/2.01E-2)*(HVOLT+5.05E-2)+4.25E1
RETURN

IF(HVOLT+2.02E-2)920,920,910

HUMDTY= -(4.4/7.04E-2)*(HVOLT+2.02E-2)+4.64El
RETURN

IF(HVOLT+4.06E-2)940,930,930

HUMDTY= -(2.6/2.04E-2)*(HVOLT+4.06E-2)+4.9El
RETURN

HUMDTY= -(l.0/l.OE-2)*(HVOLT+5.06E-2)+5.0E1
RETURN

IF(HVOLT—l.OlE-2)950,960,960

HUMDTY= -(6.3/4.05E-2)*(HVOLT-l.OlE-2)+5.58El
RETURN

IF(HVOLT+1.02E-2)970,980,980

HUMDTY= -(4.7/2.03E—2)*(HVOLT+1.02E—2)+6.05El
RETURN

IF(HVOLT+3.0E-2)990,991,991

HUMDTY= -(7.0/l.98E-2)*(HVOLT+3.0E-2)+6.75E1
RETURN

IF(HVOLT+4.05E-2)993,992,992

HUMDTY= -(7.5/1.05E-2)*(HVOLT+4.05E—2)+7.5E1
RETURN

IF(HVOLT+4.5E-2)995,994,994

HUMDTY= -(4.3/4.5E-3)*(HVOLT+4.5E—2)+7.93E1
RETURN

HUMDTY= -(9.7/5.8E-3)*(HVOLT+5.08E-2)+8.9E1
RETURN

END

BIBLIOGRAPHY

10.

11.

105

BIBLIOGRAPHY

Theodore D. Sterling, "Humanizing Computerized Information Systems."
Science, Vol. 190, 1975, pp. 1168-1172.

Dean L. Haynes, Richard K. Brandenburg, and P. David Fisher, "Envi—
ronmental Monitoring Network for Pest Management Systems," MSU
publication, Department of Entomology, 1973.

Frank J. Oliver. Practical Instrumentation Transducers. Hayden:
New York, NY, 1971, Chapters 11, 12, 13.

D. H. Sheingold, Ed. Analgngigital Conversion Handbook. Analog
Devices, Inc.: Norwood, Massachusetts, 1972.

Physical-Chemical Research Corporation. Sensor Characteristics and
Specifications pamphlet supplied by manufacturer of sensor., 1973.

COS/MOS Integrated Circuits Manual. RCA Semiconductor publication,
Somerville, New Jersey, 1975.

McMos Handbook (First U.S. Edition). Motorola Semiconductor publi-
cation, Phoenix, Arizona, 1974.

34000 Isoplanar CMOS Data Book. Fairchild Semdconductor publication,
Mountain View, California, 1975

Burr-Brown. Operation Amplifiers: Desigp and Applications. McGraw-
Hill: New York, N.Y., 1971.

Burr-Brown. Applications of Operational Amplifiers: Third Genera-
tion Techniqpes. McGraw-Hill: New York, N.Y., 1973.

Siliconex. Applications of Triple Operational Amplifier IC, L144A.
Pamphlet published by manufacturer.

||I|
llll
I|I|