Middleton W.M. (ed.) Reference Data for Engineers: Radio, Electronics, Computer and Communications
Подождите немного. Документ загружается.
12-32
-
Y
Y
2
-45-
3
w
-40-
z
-50-
REFERENCE
DATA
FOR ENGINEERS
If the noise load is removed, the thermal noise
component
in
the measurement channel can be mea-
sured. (Thermal noise is sometimes referred to as basic
noise or intrinsic noise.) Then if the noise load is
added, the additional noise introduced by intermodula-
tion
and cross talk can be measured.
By varying the level of the baseband noise loading
signal, it is possible to explore the combined effects of
thermal noise and intermodulation noise to determine
the optimum operating point for the system.
The NPR curve is often referred to as a
“V”
curve
because
of
the characteristic shape (Fig. 71). At low
loading levels, thermal noise will predominate, and
NPR will simply be proportional to the loading power
level. Thus this part of the characteristic is linear. As
intermodulation noise rises, however, the noise power
ratio will fall rapidly and give the shape of Fig.
71.
It is
possible to gain troubleshooting information by inspect-
ing the
V
curve for slots placed at different frequencies
throughout the baseband. The method employed is to
select a slot and observe the change in NPR for a I-dB
change
in
noise loading.
In
the linear region of the
curve, a 1-dB change
in
load will give a 1-dB change in
NPR. For the nonlinear segment of the
V
curve, the
change in NPR can then be used to determine the order
of the nonlinearity involved.* For example, a 2-dB
change in NPR will indicate that third-order nonlineari-
ty is the cause.
The conventional form of test set? for carrying out
NPR measurements is shown in Fig.
72.
Fixed filters
are used at both transmit and receive points in the
system. This form of test set has the advantage that
-
*
References 24 and 25.
t
Reference 24.
INTERMODULATION
NOISE
SYSTEM
LOAD
IdBmO)
Fig.
71.
NPR
curve
for
a multichannel radio system.
intermodulation in its receive mixer is not important
since bandpass filters reduce the noise band presented
to the mixer. Unfortunately, it is necessary to have a
range of bandpass filters and oscillators to cope with all
the slots and systems that may have to be measured.
The modern design of highly linear mixers, however,
allows a selective receiver to make the NPR measure-
ment. Such receivers use a synthesizer as the local
oscillator,
so
the requirement for separate filters and
oscillators is avoided.
Conversions-It is sometimes necessary to convert
from NPR to other measures, such as signal-to-noise
ratio (SNR). The following relations are used.
1.
Weighted signal-to-noise ratio in decibels:
Weighted SNR
=
NPR
+
10
log10(f/3.1)
-
P+W
where,
f
=
bandwidth of baseband
(kHz),
P
=
nominal load (dBmO),
W
=
weighting factor (dB).
HIGH-PASS LOW PASS
FILTER FILTER
NOISE LOAD
SIGNAL
NOISE
SLOT OUT
I
SOURCE
I
I
-
I
L
SLOT
NOISE
BANDWIDTH
BANDPASS
FILTER
Fig.
72.
NPR
measurement
12-33
When applied to systems having greater than
240
channels and using C-message weighting, this
simplifies to
SNR
=
NPR
+
17.8
2.
The notation dBrnCO means decibels of noise
with respect to reference noise of
-90
dBm and
with
C
message weighting, corrected
to
a point of
zero transmission level. Conversion to dBrnCO for
systems that have more than
240
channels is given
by
dBrnCO
=
71.7
-
NPR
Note that the European psophometric weighting
and
C
message differ by
I
dB, but in practice the
above formula is often used for psophometric
weighting.
Carrier/Noise Measurement
For digital radio systems, the normal overall perform-
ance measurement is based on “bit error rate,” where a
pseudo-random-sequence bit pattern is applied to the
system and errors are indicative of system imperfec-
tions.
A
more detailed analysis is made possible by
inspection of the so-called “eye” diagram at the
sampling stage of digital signal recovery, but measure-
ment of error rate as the carrier-to-noise ratio is varied
can be a more revealing method of analyzing system
performance.
Diagnostic Measurements and
System Performance
The relationships between diagnostic tests and system
performance are often very complex. Thus it is difficult
to relate measurements of envelope-delay distortion to
the intermodulation that will be produced. It is possible
to make such predictions,* but this generally requires
the use of a computer. It is generally more useful to
derive a clear understanding of how these measurements
can best be used to optimize system performance,
or
at
least to be able to separate those measurement respons-
es that have a significant effect on system performance
from those that have a lesser effect.
COMPUTER CONTROL
OF
INSTRUMENTS
A
series of standards provides mechanisms for com-
puters to communicate with instruments. IEEE 488.1
or
IEC 625-1 describes a physical interface including
electrical and mechanical constraints. IEEE 488.2
or
IEC 625-2 describes how complete messages are for-
matted and exchanged. SCPI defines the semantics of
the messages.
*
Reference
17.
IEEE
488.1
or IEC
625-1
General Purpose
Interface Bus (GPlB)*
The GPIB provides a means by which system compo-
nents communicate with each other. It
is
analogous to
the telephone system that allows one person
to
exchange
verbal information with another person-or,
in
the case
of a conference call, with several other people.
Every participating device in a GPIB system per-
forms at least one
of
three roles: talker, listener,
or
controller.
A
talker transmits data to other devices via
the bus.
A
listener receives data from other devices.
A
controller
n-
3ges communications on the bus, primar-
ily by desigilating which devices are to send
or
receive
data during each measurement sequence. The controller
may also interrupt and command specific actions within
devices.
Many devices are both talkers and listeners.
For
example, a programmable multimeter
or
electronic
counter listens when receiving its program instructions,
and it talks when sending its measurements
to
another
system component such as a printer
or
computer. There
can be several listeners, but to avoid confusion, only
one talker can be active on the bus at any one time.
In
its simplest form, a GPIB system can consist
of
only one talker and one listener-for example,
an
electronic counter linked to a printer. In this applica-
tion, there is no need for a controller; the counter
“talks” and the printer “listens.”
The full versatility of the GPIB becomes apparent
when there is more stimulative interaction among the
interconnected devices. To achieve this requires a con-
troller that can schedule measurement tasks, set up
instruments to perform specified tests and measure-
ments, monitor processes on-line, and process data and
analyze and interpret the results.
A
desktop computer,
or
“computing controller.”
is
often used as a system
manager to control the bidirectional data flow of the
GPIB.
Interface circuitry is part of each interfaced instru-
ment
or
computing controller-the GPIB cable itself is
passive. Inside the cable,
16
active signal lines are
grouped into three sets, according to function (Fig. 73).
The eight data lines carry coded messages-such as
addresses, program data, measurements, and status
bytes-among as many as
15
devices interconnected
with a single bus, The same data lines are used for both
input and output, and the messages are in bit-parallel,
byte-serial form. Data are exchanged asynchronously,
for maximum compatibility among a wide variety of
devices.
For unambiguous, intelligible communication be-
tween instrument and computer devices, some rules
or
protocol must apply to the communication process
itself. Thus, the exchange
of
data is controlled by the
*
From
Hewlett-Packard publication
5021-1927.
0
Hew-
lett-Packard, used
with
permission.
REFERENCE
DATA
DEVICE A DEVICE
B
DEVICE C
ABLE
TO
TALK
ABLE
TO TALK
ABLE
TO
USTEN
USTEN AND AND USTEN ONLY
(e.g.
calculator) (e.g. digital voltmeter) (e.g. signal generator)
CONTROL
FOR
ENGINEERS
DEVICE D
ABLE
TO
TALK
(e.g. tape recorder)
EO1
REM
SRQ
Am
IFC
NDAC
NRFD
DAV
DIO
1..
8
GENERAL
BUS
HANDSHAKE DATA
BUS
MANAGEMENT UNES
Fig.
73.
General purpose interface
bus.
(From Hewlett-Packard publication
5021
-1927.
0
Hewlett-Packard, used with permission.)
second set of signal lines, the three data byte transfer
control lines.
All
listeners addressed on the GPIB send
a signal when they are “ready for data.” The talker then
initiates data-byte transfer by signaling that “valid
data” are on the data lines. When each listening device
successfully receives the message, it acknowledges
“data accepted.”
When two or more devices are listening, all must
acknowledge “data accepted” before new data can be
transmitted on the bus. Thus, data are transferred at the
rate of the slowest listener participating in the particular
conversation.
High-speed and slow devices can be connected on the
same bus. Data transfer between high-speed devices is
not adversely affected, as long as the slow devices are
not addressed or participating in the conversation.
The remaining five general interface management
lines are used for such things as activating all the
connected devices at once, clearing the interface, and
so
forth.
Up to
15
GPIB devices can be linked together on one
bus. The GPIB cable connects them in parallel func-
tionally, in either star or linear fashion physically. Total
cable transmission path length on any one
bus
should
not exceed
20
meters or
2
meters per device, whichever
is less, except where the distance is extended by devices
such as a common carrier interface.
A single computing controller can manage more than
one bus instrument cluster at once. This is particularly
useful for achieving concurrent actions at different test
stations in a production test measurement application.
Interface circuitry for controllers and instruments is
commercially available
in
integrated-circuit packages.
IEEE
488.2
or
IEC
625-2
As illustrated in Fig. 74, IEEE 488.2 is a higher level
standard than GPIB. It has also been used on top of
IEEE
1155,
VME Extensions for Instrumentation
(VXI).
While the physical interface provides a connec-
tion between system components, IEEE 488.2 provides
rules on how the communication
on
the interface will
take place.
The major areas covered by IEEE
488.2
are:
1.
2.
3.
4.
5.
Reliable transfer of complete messages between a
computer and an instrument.
A
precise description of the syntax in those
messages.
The form and meaning of a set of commands
useful in all instruments.
Common status reporting capability.
Techniques for guaranteeing synchronization of
application programs with instrument functions.
IEEE 488.1 provides a reliable means of transferring
bytes between a talker and a listener.
In
addition, a
reliable means of transferring programming commands
and measurement results is needed.
A
response is generated only after a query message
has been received. An instrument responds only with
the data asked for by the controller. An instrument is
not allowed to generate arbitrary data when it is
addressed to talk. Any violations of the message ex-
change protocol are reported as query errors in the
event-status register described by the status reporting
model.
IEEE
488.2
requires that devices listen in a “forgiv-
ing” manner. The level of forgiveness is, however, very
precisely defined. The syntactic meaning of every data
byte is well defined, even when multiple characters have
the same meaning.
While the listening syntax may be very forgiving, an
instrument is required to talk very precisely. As a result,
the chances of a message being accepted by any
controller are greatly increased.
Certain functions are useful across all types of
instruments.
IEEE 488.2 defines a set of commands
which all instruments must implement exactly as de-
scribed. Some of the more generic commands are
resetting and identifying the instrument. Other com-
mands are described which may be useful in some
instruments. They are included in the standard
so
that if
the function
is
implemented, implementation will be
done
in
a standard way.
IEEE 488.2 describes a status-reporting model that is
hierarchical. The contents of other data structures are
MEASUREMENTS AND ANALYSIS
12-35
+
SYSTEM
COMPONENTx
MFRS.
:
lEFE488.2
SPECS. STANDARD
;
lEEE
488.1
STANDARD
+
SYSTEM
*
COMPONENTy
WHERE:
Layer D represents Device Functions.
Layer C represents Common System Functions.
Layer
B
represents
Message
Communication Functions.
Layer
A
represents Interface Functions
(IF).
:
lEEE488.2
:
MFR5
:
STANDARD SPECS.
Fig.
14.
IEEE
488.2
layers.
(From Hewlett-Packard publication
5021-1927.
0
Hewlett-Packard, used with permission.)
summarized in the status byte. Some of the conditions
required are whether a message is available and whether
an error has occurred.
Instruments are often able to accept commands faster
than they can be executed. While this feature can
improve system throughput, it also creates a need to
know when the instrument has actually completed all its
commands.
IEEE
488.2
requires commands which
enable the application program to detect when all
pending operations are complete.
Standard Commands
for
Programmable Instruments
(SCPI)
SCPI
fills
in the top layer of Fig.
74
by describing the
semantics
of
the messages. The most important objec-
tive
of
SCPI
is to be consistently usable with a large
variety
of
instruments. This horizontal compatibility
means that a power supply can be programmed similarly
to a microwave network analyzer. Even though these
two instruments have very different capabilities, there
is
no reason to use two different programming languages.
Although having a consistent style is convenient, the
ability
to
use the exact same command for the same
function (even with different hardware) is more impor-
tant. Thus, a voltmeter uses the same command as an
oscilloscope for measuring dc voltage.
This compatibility is derived from using a genera-
lized model of an instrument. This model, along with
the underlying standards, enforces a discipline on the
language designers. In an effort to meet the needs of
advancing instruments needs, a consortium of instru-
ment manufacturers manages the growth
of
the lan-
guage.
ELECTROMAGNETIC
COMPATIBILITY,
INTERFERENCE,
AND
SUSCEPTIBILITY
EMC/EMI/EMS
Electromagnetic compatibility
(EMC)
is the ability
of items of electronic equipment to function properly
together in the electronic environment. Most electronic
12-36
CHART
I.
TYPICAL
EMUEMS MEASUREMENTS
Unwanted Emissions Unwanted Susceptibility
Conducted Conducted
Power Lines Power
Lines
Signal lines Continuous wave
Control lines Dropouts, surges
Antenna leads Fast transients
Electric fields Control lines
Magnetic fields Receiving antenna
Radiated Signal lines
Radiated
Electric Field
Magnetic field
Electrostatic Discharge (ESD)
equipment is designed to perform an assigned task.
However, when improperly designed
or
poorly main-
tained, a device may emit unintentional radiation to
interfere with the proper functioning
of
other equip-
ment. Sometimes equipment is designed to emit radia-
tion at designated frequencies and with prescribed
bandwidths but may inadvertently emit at other frequen-
cies or harmonics. These situations create electromag-
netic interference (EMI) to other equipment.
Engineers would prefer
to
design equipment to sur-
vive under the most severe electromagnetic interfer-
ence,' but they are often constrained by size, weight,
and cost requirements. Designing equipment to work at
prescribed interference levels is designing for accepta-
ble susceptibility (EMS).
Regulations
In the United States, the Federal Communications
Commission is the body responsible for the control of
EMI. Part
15
(for radio-frequency devices) and Part
18
(for medical, scientific, and industrial equipment)
of
the FCC Rules and Regulationst contain sections
on
the
control
of
interference. Devices intended for commer-
cial and scientific applications with limited usage are
governed by class A limits. Devices intended for home
or consumer use are covered by class
B
limits, which are
up to
22
dB lower in level of conducted emission and
10
dB
lower in level
of
radiated emission. Certification
from the FCC
is
required before equipment may be sold
in the USA.
Another source of EM1 regulations, particularly for
builders of military equipment, is MIL-STD-461. Test
methods for the frequency range
30
Hz
to 40
GHz
are
covered in MIL-STD-462.
European Community countries use unified EMC
standards
(EN)§
that regulate the level of permissible
electromagnetic emission and also susceptibility. These
standards are based
on
IECS (International Electrotech-
nical Commission) and CISPRP (International Special
Committee on Radio Interference) recommendations.
Certification
to
these regulations must be performed
before the equipment is offered for sale.
Most countries in Asia and other parts of the world
follow the CISPR recommendations in the development
of their national EM1 standards.
Other international bodies, such as the ITU (Interna-
tional Telecommunication Union), provide detailed
limits of EM1 for specialized equipment. In general,
compliance with their standards is voluntary, but com-
pliance will often mean better chances of success in
interfacing with existing equipment.
Typical emission limits are shown in Fig.
75.
EMI/EMS
Measurements
Chart
1
lists typical EMIiEMS measurements. An
open area test site** (OATS) with little
or
no electro-
magnetic ambient noise is ideal
for
making emission
measurements. An anechoic chamber, a shielded room
with absorbent material lining all walls, is ideal for
susceptibility measurements. Emission measurements
in a semianechoic chamber with a conductive floor as
ground plane may substitute for OATS measurements
within limitations. Unlined screen rooms are afflicted
with multiple reflections, which add
to
or
subtract from
the true magnitude of the radiation, and are useful only
for conducted-EMC measurements.
Inexpensive techniques for the testing of radiated
EMC are offered by TEM cellstf (transmission lines) if
-
*
Ott, Henry W.,
Note Reduction Techniques in Electronic
Systems.
New York: Wiley-Interscience.
Code of Federal Regulations, Title
47.
Washington, DC
20402:
US
Government Printing Office.
§
EN,
IEC,
and CISPR standards may
be
purchased from:
Sales Department, American National Standards Institute,
1
1
West 42nd Street, New York, NY 10036.
*
*
American National Standard Guide for Construction of
Open Area Test Sites
for
Performing Radiated Emission Meas-
urements,
ANSI
C63.7-1988. (IEEE, Inc., 345 East 47th
Street, New York,
NY
10017.)
t?
Crawford, M.
L.,
Workman,
J.
L.
Using a TEM Cell for
EMC Measurements
of
Electronic Equipment,
NBS
Technical
Note 1013, 1981. Boulder, CO
80303:
NIST.
MEASUREMENTS AND ANALYSIS
12-37
3
-.
2
t
12-38
REFERENCE DATA FOR ENGINEERS
the equipment
is
small relative to the dimensions of the
cell. For EM1 measurements, a receiver is connected as
the load to the cell, and for EMS measurements, a
signal generator
is
connected as the source. The TEM
cell is completely shielded. As the cell dimensions
become comparable to the wavelength, higher-order
modes will affect the calibration thus creating a high-
frequency usage limit known as the cutoff frequency.
The cutoff frequency is raised considerably
in
a special
form of the TEM cell* where the voltage and current
on
the transmission line are internally terminated and the
associated electromagnetic field is absorbed.
Strip-line cells? are sometimes used for testing small
electronic modules. They are inexpensive, easy
to
build, and economical of bench space. Unfortunately,
they are not shielded, and their use is therefore limited
to preliminary testing and troubleshooting.
Emission sources may be located during trouble-
shooting with inexpensive, small hand-held probes$ for
magnetic fields. Such probes may also be used to inject
fields for localized susceptibility tests.
REFERENCES
Error Correction
1.
2.
3.
4.
5.
Kruppa, W., and Sodomsky,
K.
F. “An explicit
solution for the scattering parameters of a linear
two-port measured with an imperfect test set.”
IEEE Trans. Microwave Theory Tech.,
Vol. MTT-
19, Jan. 1971, pp. 122-123.
Speciale,
R.
A.
“A generalization of the TSD
network-analyzer calibration procedure, covering
n-port scattering-parameter measurements, affect-
ed by leakage errors.”
IEEE Trans. Microwave
Theory Tech.,
Vol. MTT-25, Dec. 1977, pp.
Fitzpatrick,
J.
“Error models for systems measure-
ments.”
Microwave
J.,
No.
5,
1978, pp. 63-66.
Engen, 6. F., and Hoer, C. A. “Thru-reflect-line:
An
improved technique for calibrating the dual
six-port automatic network analyzer.
”
IEEE
Trans. Microwave Theory Tech.,
Vol.
MTT-27,
Apr. 1979, pp. 987-993.
Soares, R. A., et al. “A unified mathematical
approach
to
two-port calibration techniques
and
some
applications.
”
IEEE
Trans. Microwave The-
ory Tech.,
Vol. 37, Nov. 1989, pp. 1669-1674;
also Vol. 38, Aug. 1990, pp.
1144-1
145.
E
100-1 1
15.
-
*
Wilson,
P.,
Hansen, D., Koenigstein,
D.
“Simulating
Open
Area
Test
Site Emission Measurements Based on Data
Obtained
in
a Novel TEM Cell,’’ IEEE 1989 International
Symposium
on
EMC, IEEE Cat.
No.
89CH2736-7,
pp.
I71 -177.
t
Engineering
Design
Handbook-EMC.
DARCOM-P 706-
410
(US
Army;
pp.
7-49).
5
Terrien,
M.
“EM1 Troubleshooting Techniques Utilizing
a Broadband Magnetic Field Sensor.
”
International Zurich
Symposium
on
EMC, 1987.
6.
Eul,
H. J., and Schiek,
B.
“A
generalized theory
and new calibration procedures for network analyz-
er self-calibration.
”
IEEE Trans. Microwave The-
ory Tech.,
Vol. 39, Apr. 1991, pp. 724-731.
7.
Eul,
H.
J.,
and Schiek,
B.
“Reducing the number
of calibration standards for network analyzer cali-
bration.”
IEEE Trans. Instrum. Meas.,
Vol. 40,
8. Butler, John V.; Rytting, Douglas K.; Iskander,
Magdy F.; Pollard, Roger D.; and Vanden Bossche,
Marc.
‘‘
16-term error model and calibration proce-
dure for on-wafer network analysis measure-
ments.”
IEEE Trans. Microwave Theory Tech.,
Vol. 39, No. 12, Dec. 1991, pp. 2211-2217.
9. Silvonen, Kimmo J. “A general approach to
network analyzer calibration.
”
IEEE Trans. Mi-
crowave Theory Tech.,
Vol.
40,
No.
4, Apr. 1992.
Aug. 1991, pp. 732-735.
Time and Frequency
Measurement
10.
Barnes,
J.
A., et al. “Characterization of Frequen-
cy Stability.”
IEEE Trans. Instrum. Meas.,
Vol.
IM-20, No.
2,
May 1971, pp. 105-120.
11.
“Standard Frequencies and Time Signals.”
Vol.
VII, CCIR
XI11
Plenary Assembly. Geneva: Inter-
national Telecommunication Union, 1975, p. 18.
12. Fischer, Michael C. “Frequency Stability Mea-
surement Procedures.
”
Proceedings, Eighth An-
nual PTTI Meeting, Goddard Space Flight Center,
Code 250, Greenbelt, Md., 1976, pp. 575-618.
13. Fischer, Michael C. “Analyze Noise Spectra With
Tailored Test Gear.”
Microwaves,
Vol.
18,
No. 7,
July 1979, pp. 66-75.
14.
Fundamentals
of
the Electronic Counters,
Appli-
cation Note 200.
Palo
Alto, Calif.: Hewlett-
Packard
,
1978.
15.
Fundamentals
of
Time and Frequency Standards,
Application Note 52-1. Palo Alto, Calif.: Hewlett-
Packard, 1974.
16. Kartaschoff, Peter.
Frequency and Time.
New
York: Academic Press, Inc., 1978.
17. Manassewitsch, Vadim.
Frequency Synthesizers
Theory and Design.
2nd ed. New York: John Wiley
&
Sons, Inc., 1980.
18.
Oliver,
B.
M.,
and Cage,
J.
M.,
eds.
Electronic
Measurements and Instrumentation.
New York:
McGraw-Hill
Book
Co.,
1971, Ch. 6.
19. Rutman, Jacques. “Characterization
of
Phase and
Frequency Instabilities in Precision Frequency
Sources: Fifteen Years of Progress.
”
Proceedings
of
the
IEEE,
Vol. 66, No.
9,
Sept. 1978, pp.
1048- 1075.
20.
Timekeeping and Frequency Calibration,
Applica-
tion Note 52-2. Palo Alto, Calif., Hewlett-
Packard
,
1979.
21.
Vessot, Robert F. C., et
al.
“Research With a Cold
Atomic Hydrogen Maser.
”
Proceedings 33rd Fre-
quency Control Symposium.
Washington,
D.C.:
MEASUREMENTS
AND
ANALYSIS
12-39
Electronic Industries Association, 1979, pp.
51
1-
514.
22.
Winkler, Gernot M. R. “A Brief Review of
Frequency Stability Measures.
”
Proceedings,
Eighth Annual PTTI Meeting, Goddard Space
Flight Center, Code 250, Greenbelt, Md., 1976,
pp. 489-527.
Microwave- Lin
k
Analysis
23. Differential Phase
&
Gain at Work, Application
Note 175-1. Palo Alto, Calif.: Hewlett-Packard,
Nov. 1975.
24. Tant,
M.
J.
The White Noise Book. White Crescent
Press Ltd., 1974.
25. “An Integrated Test Set for Microwave Radio Link
Baseband Analysis.” Hewlett-Packard Journal,
April 1982.
26. Smith, Emerson C. Glossary
of
Communications.
Telephony Publishing Corp., 197
1.
27. Preston, C. L. Broadband Transmission System, a
volume from the series Noise Analysis Theory and
Technique. New York: American Telephone and
Telegraph Corp., 1972.
13
Magnetic-Core
Transformers and
Reactors
Revised
by
Charles
F.
Hempstead,
Dunvood
R.
Kressler, Richard
W.
Avery,
Fred
J.
Banzi, Bernard
D.
Carniglia, Arthur Olsen,
Jr.,
Kent
W.
Sternstrom,
and
Richard C. Walker
Introduction
13-3
Definition of Transformer and Inductor
Transformer Types and Frequency Ranges
Inductor Types and Frequency Ranges
Generalized Equivalent Circuit for a Transformer
Power Transformers
13-3
Types of Magnetic Cores
Design of Power Transformers for Rectifiers
Effect of Duty Cycle
on
Design
Methods of Winding Transformers
Dielectric Insulation and Corona
Temperature and Humidity
Ferroresonant Transformers
13-10
ConvertedInverter Transformers
13-11
Audio-Frequenc y Transformers
Types of Magnetic Cores
Design
of
Audio-Frequency Transformers
13
-I
2
13-1