This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
After reading tb/s at-tide
and taking the test, thereader ui!!:
Abbreviations: AC alternating current, DC = direct current
Index terms: Physics . Radiography, technology
RadloGraphics 1997; 17:1533-1557
1533
The AAPM/RSNA PhysicsTutorial for Residents
X-ray Generators1This article meets the
criteria for 1. 0 credit
hour in Category 1 of
the AMA Physician’s
Recognition Award.
To obtain credit, see
the questionnaire on
pp 1527-1532.
. Understand the general
operational aspects and
functional components of
the x-ray generator.
. Be able to describe dcc-
tromagnetic induction prin-
ciples, basic electrical corn-
ponents, and operation of
transformers.
. Understand the differ-
ence between filament cur-rent and tube current, the
space charge effect, and its
impact on tube current.
S Be able to describe the
different types of x-ray gen-
erators and the conse-
quences of voltage wave-
form on x-ray quantity and
quality.
j Anthony Seibert, PhD
The x-ray generator delivers the electrical power to energize the x-ray tubeand permits the selection of x-ray energy, x-ray quantity, and exposure time.Major internal components of the generator include transformers, diodes andrectifier circuits, filament and stator circuits, timer switches, and kilovolt andmiffiampere meters. Single-phase, three-phase, high-frequency, and constant
potential generators produce different voltage waveforms (ripple) and x-raybeam spectra. Phototimer and automatic brightness control subsystems mea-
� sure radiation exposure incident on the image receptor to give instantaneous
j � feedback for optimal radiographic film densities and fluoroscopic imagebrightness, respectively. At the generator control console, the operator sets
the tube voltage, tube current, exposure time, phototimer film density, spotfilm acquisition, and fluoroscopic parameters. Selection of generator power
and options depends on the intended clinical use. X-ray tube focal spot sizeand power loading capability should be matched to the x-ray generator and
clinical imaging requirements. Single and multiple exposure rating charts aswell as anode and housing thermal characteristic charts indicate power input
� and dissipation rates specific to a generator and x-ray tube target and housing.
. INTRODUCTIONThe x-ray generator delivers the power to the x-ray tube necessary to produce x rays
in a defmed and predictable manner. The generator also provides mechanisms to se-
lect techniques appropriate for a given examination and to protect the x-ray tube and
patient from possible overload situations.
The objectives of this article are to provide an overview of the various components
that constitute the x-ray generator, to illustrate their functions, and to explain the key
roles of the x-ray generator in the production of x rays. Principles of electromagnetic
induction, generator design, electronic circuitry design, characteristics of the high-
voltage waveform, methods of exposure control, power specifications, tube rating
charts, and anode and housing thermal characteristic charts are explained. Much of
the information contained herein is attributed to several excellent references (1-6).
. Be aware of the issues re-lated to generator power,tube loading limitations,
and rating charts.
‘From the Department of Radiology, School of Medicine, University of California Davis Medical Center. FOLB-2E, 2421
45th Street, Sacramento, California 95817. From the AAPM/RSNA Physics Tutorial at the 1996 RSNA scientific assembly.
Received May 23, 1997; revision requested July 17 and received August 19; accepted August 2 1 . Address reprint re-
Reverse bias: Forward bias:No electron flow Electron flow
� I +++++� � � i � I ++:� � l___J ‘� +++++
I +++++ � I �
e- � �c�:Z�:::= e�
November-December 1997 Seibert U RadioGraphics U 1539
Figure 9. Circuit diagram symbols for
valve and solid-state diodes are shown. In
each case, flow through the diode fromthe electron source to the electron target.The solid-state diode is represented witha triangle and a connecting straight verti-
cal line with the apex of the triangle,
where electrons flow in the direction
passing the vertical line and then through
the triangle. Although this symbol iscounterintuitive when one views the di-
ode as an “arrow, ‘ current has historically
been described as the flow of positivecharge (ie, opposite the electron flow di-
rection). In this context, the diode actu-ally does point in the direction of current
(flow of positive charge). Because dee-trons flow through the circuit oppositethe direction of the current, the diode (ifviewed as an arrow) points in the block-
ing direction of electron flow. Bottom
diagrams show the design and operation
of a solid-state diode N-P junction. When
a positive charge is placed on the P-type
material and negative charge on the N-
type material, a one-way flow of currentoccurs through the junction. Note the di-rectional flow of electrons through the di-
ode device from a wmbolic viewpoint.
pentode (electrical devices with three, four,
and five electrodes, respectively). A diode con-
tains an electron SOUrCe (cathode) and an elec-
tron target (anode) to permit current to flow in
only one direction along a conductor, despite
any changes in voltage polarity in the electrical
circuit. A triode contains a cathode, anode, and
a third control electrode to adjust or switch
(turn on and off) the current. Tetrodes and
pentodes contain additional control electrodes
but function similar to triodes. In general, these
devices are constructed with vacuum or gas-
filled envelopes or solid-state materials.
The valve diode, which contains an anode
and cathode within a vacuum envelope, per-
mits the flow of electrons only when the polar-
ity of the cathode is negative and the anode
positive. A pertinent example is the x-ray tube,
in which electrons flow in the circuit from the
heated cathode (the source of electrons cre-
ated by thermionic emission) to the positive
anode. When the polarity of the anode and
cathode reverses during the AC cycle, current
ceases because no electrons are available from
the anode to flow to the cathode. Thus, cur-rent is permitted only when the electron
source and target have the correct polarity and
is blocked in the other direction.
Solid-state diodes are composed of semicon-
ductors (materials such as pure silicon that
conduct electricity under certain conditions)
doped with impurities (eg, boron, indium, alu-
minum, gallium) that provide a surplus of free
electrons in N-type materials (a net negative
charge) or a shortage of electrons in P-type ma-
terials (a net positive charge). These N-typeand P-type semiconductors are joined together
in such a way as to permit electrons to pass tin-
inhibited in one direction when the potential
difference is applied negatively and positively
to the N-type and P-type materials, respec-
tively. In this situation, the electron surplus in
the N-type material and the electron shortage
in the P-type material are simultaneously in-
creased, and electrons are conducted across
the junction. If the voltage bias is reversed,
however, the electron surplus in the N-type
material and the electron shortage in the P-type
material are both decreased, which causes a
depletion layer to form at the junction and pro-
hibits electron flow within the semiconductor
materials (similar to the preceding analogy of
the x-ray tube). Design and operation of the
solid-state diode P-N junction under reverse
and forward bias are shown in Figure 9.
1540 U Imaging & Therapeutic Technology Volume 17 Number 6
Figure 10. Diagrams show a
vacuum-based triode switch (left),its electronic symbol (center), andthe circuit diagram symbol for asolid-state triode device (a thyristor)
(right).
The triode is a diode with the addition of a
third electrode placed close to the cathode (Fig
10). This additional electrode, called a grid, is
situated between the cathode and anode. All
electrons must therefore pass through the grid
structure. Electrons flow freely in the triode
when the grid has no applied voltage; how-
ever, because the grid is much closer to the
cathode than the anode, a small negative volt-
age can exert a tremendous force on the elec-
trons emitted from the cathode and effectively
stop the current. Even with extremely heavy
loads applied on the secondary side of the x-
ray transformer, in a properly confIgured tri-
ode, the current can be rapidly switched sim-
ply by controlling the grid voltage by placing a
nominally small negative voltage on the grid.
A “grid-switched” x-ray tube is a notable ex-
ample of a triode; in this device, the cathode
cup is isolated from the filament structure and
negatively biased to turn the x-ray tube current
on and off with microsecond accuracy. Gas-
filled triodes that contain thermionic cathodes
are known as thyratrons. Solid-state triodes,
known as thyristors, operate with the use of a“gate” electrode at the N-P junction. These
switches control the timing of the x-ray expo-
sure under heavy power loads.
Tetrodes and pentodes are also used in
some x-ray generators. These devices have
multiple grid electrodes, which provide further
control of the current and voltage characteris-
tics. Thermionic diodes, triodes, tetrodes, and
pentodes have almost universally been re-
placed by their solid-state counterparts in mod-
em x-ray generators.
Anode
Grid __,/JCathodeelectrode
. Rectifier Circuits
Rectifier circuits are composed of two or more
diodes arranged to divert the flow of electrons
through a single path of a multipath circuit. A
bridge rectifier is an electrical circuit com-
posed of multiple diodes on the secondary side
of the high-voltage transformer in an x-ray gen-
erator. The bridge rectifier diverts electron
flow through the high-voltage circuit to ensure
correct polarity of the cathode (negative) and
anode (positive) at all times.
For example, during the first (positive) half-
cycle of the AC waveform, one side of thetransformer is maintained at one-haLf the posi-
tive peak voltage, and the other side is main-
tamed at one-half the negative peak voltage
(Fig 1 1). Electrons flow from the negative to
the positive side of the circuit. Through one
side of the bridge rectifier, the electrons pass
through only one diode, as two others actively
block their flow. Thus, the electrons are routed
through the cathode of the x-ray tube to the an-
ode and back into the bridge rectifier circuit.
At this juncture, any one of three diodes in the
rectifier circuit will allow the electrons to pass,but only the diode directly connected to the
positive pole of the transformer will conduct.
In the next half-cycle, the polarity of the trans-
former is reversed. The electrons flow from
the negative pole (opposite side of the trans-
former from the previous half-cycle), through
the bridge rectifier through only one diode, to
the x-ray tube, and back to ground (Fig 1 1). For
this half-cycle (reverse polarity), the other two
diodes pass electrons through the circuit.
Therefore, the applied voltage across the x-ray
tube always has the correct polarity, and a rec-
tified AC waveform with two lobes per cycle is
produced.
:�+
First half-cycle
November-December 1997 Seibert U RadioGraphics U 1541
X-ray tube
Second half-cycle
Figure 11. Diagram illustrates the
flow of electrons through a bridgerectifier circuit during each half-cycle of the applied voltage. Thegroup of diodes is arranged so that
the electron flow is always divertedthrough the cathode to the anode ofthe x-ray tube, independent of thechange in polarity of the high-volt-age transformer. For each half-cycle,two of the four diodes conduct(black diode symbols).
Figure 12. Diagram shows the components of a filament circuit and its relationship to thex-ray tube. A space-charge compensation circuit is present in the filament circuit to correct
for nonlinear attributes of the filament current relative to the x-ray tube current, particularly
when the generator is operated below 40 kVp.
I Filament Circuits
The filament circuit consists of a step-down
transformer connected by output lines to the
focal spot filaments in the cathode of the x-ray
tube (Fig 1 2). When voltage is applied to the x-
ray tube, a current passes through the selected
filament: This is the filament current. Electrical
resistance causes the filament to heat up and
release electrons by thermionic emission. The
number of electrons released is related to the
filament current: The higher the current, the
higher the heat and the greater the number of
electrons released. When a high voltage is ap-
plied to the x-ray tube, all free electrons re-
leased from the filament surface are acceler-
ated toward the anode. This flow of electrons
constitutes the tube current, which is the rate
of electron flow from the cathode to the an-
ode. Thus, tube current is controlled by but not
the same as filament current.
Filament current (1 - 10 A) is much greater
than tube current (1-1,000 mA), usually by a
factor of 10 or more. The resultant tube current
is proportional to the ifiament current except
when the tube is operated at 40 kVp or less.
The filament current is adjusted through a Se-
ries of resistors (or a variable rheostat) to con-
trol the heating and subsequent thermionic
emission of electrons.
- C,)� .�
C)�-.
�C-)>
0)
current
0 10 20 30 40 50 60 70 80 90 100 kVp
Applied tube voltage
1542 U Imaging & Therapeutic Technology Volume 17 Number 6
Figure 13. Graph depicts the rela-
tionship of the filament current to
the tube current as a function of
tube potential. The three curves il-lustrate the nearly proportional rela-
tionship between filament current
and tube current above 40 kVp,
where emission limited output oc-curs. Below 40 kVp. the tube cur-
rent becomes space charge limited,
resulting in a reduced tube current
for a specific filament current. (Re-
drawn, with permission, from refer-
ence 6.)
Space charge is the accumulation of dcc-
trons around the filament before high voltage is
applied to the x-ray tube. Like-charge repulsion
of the electrons creates a coulombic force that
quickly equals the force of thermionic emis-
sion, creating a steady-state transfer of dcc-
trons to and from the filament surface. Space
charge has an important impact on the relation-
ship between filament current and tube cur-
rent, particularly when an x-ray tube is oper-
ated at tube potentials below 40 kVp (Fig 13).
At tube potentials above 40 kVp, which is
referred to as emission-limited operation, fila-
ment current is proportional to x-ray tube cur-
rent. Here, high, medium, or low filament cur-
rent results in a proportional high, medium, or
low tube current, essentially independent of
the applied tube potential. Filament current of
3 A and higher (�3,000 mA) is typical, whereas
the tube current ranges from about 50 to 800
mA for general diagnostic work performed
with single exposure techniques. Only minor
compensation for variations in filament and
tube current is necessary for variations in x-ray
tube voltage. At tube potentials below 40 kVp,
space charge-limited operation occurs; that is,
the accumulated electron cloud surrounding
the heated filament does not instantaneously
accelerate to the anode in the x-ray tube. In
these cases, because of the nonlinear relation-
ship between filament current and tube cur-
rent, space-charge compensation circuits are
implemented to adjust the filament current so
that an accurately known tube current can be
delivered.
Separate filament circuits (step-down trans-
formers) are directly connected on the second-
ary side of the circuit to the long and short fo-
cal spot filaments in the x-ray tube. When theoperator selects a large focal spot, the step-
down transformer connected to the long fila-
ment is energized; similarly, when a small focal
spot is selected, the short filament is ener-
gized.
. GENERATOR DESIGNSSeveral designs of x-ray generators of varying
complexity and cost are currently in use: the
single-phase generator, the three-phase genera-
tor, the constant potential generator, and thehigh-frequency generator. These generator
types all incorporate the following basic com-
ponents: (a) the high-voltage transformer (a
step-up transformer), which converts the low
voltage to a proportional high voltage; (b) the
filament transformer (a step-down trans-
former), which controls the filament current
and tube current; (c) the autotransformer (used
only in single- and three-phase generators),
which compensates for input line voltage
changes; (ii) rectifier circuits, which reroute
the flow of electrons in the high-voltage circuit
of the generator, ensuring the correct polarity
of the cathode and anode, independent of the
waveform polarity; (e) the kilovolt meter,
which indicates the voltage applied to the x-ray
tube and measures the potential difference on
the primary side of the high-voltage circuit to
minimize the hazards of high voltage; and
(f) the milliampere meter, which measures
the current through the x-ray tube on the high-
voltage side of the transformer.
November-December 1997 Seibert U RadioGraphics U 1543
Line voltagecompensation
220 VSinglePhase
� Input
4-
I
Figure 14. Circuit diagram of a single-phase full-wave-rectified generator illustrates the various major Compo-
nents of the generator. (Redrawn. with permission. from reference 6.)
. Single-Phase GeneratorsA single-phase x-ray generator with a simplified
exists on the other pole. The overall potentialdifference between the poles is +50 kV - (-50kV) = 100 kV. Relative to ground, however, the
Cable capacitance effects
1544 U Imaging & Therapeutic Technology Volume 17 Number 6
60 HzAC linevoltage input
Transformer
1\f\Low voltage
1W��Pulsating DCOutput voltage
kV
mA
mA is emission limited above 40 kVp
Figure 15. Diagrams illustrate the output voltage and current for a single-
phase, full-wave-rectified generator. The rectified AC output waveform var-
ies from 0 V to the peak voltage for each half-cycle (top). Tube current does
not directly follow applied voltage due to space charge effects below 40
kVp (lower left). Cable capacitance minimizes the wide swings in voltage
from pulse to pulse (lower right). (Redrawn, with permission, from refer-
ence 6.)
peak voltage is only one-half of the peak poten-tial difference, which minimizes the electrical
insulation safety requirements for the high-volt-
age cables. (For mammographic x-ray tubes,
center-tapping usually is not used, but these
tubes operate to a maximum of only about 50
kVp.)
The miffiampere meter directly measuresthe tube current in the secondary circuit. Be-
cause the meter is placed near the electrical
ground of the secondary circuit, electrical
safety is ensured, even though the meter is on
the high-voltage circuit.
The bridge rectifier circuit, which is corn-
posed of four diodes in a single-phase genera-
tor, ensures that electrons in the secondary cir-
cuit always flow from the cathode to the an-
ode, despite the changes in secondary voltage
polarity with time.
“Two-pulse” single-phase generator is thename often attributed to single-phase genera-
tors with a bridge rectifier. Self-rectified or half-
wave-rectified “one-pulse” single-phase genera-
tors do not use a bridge rectifier and produceonly one pulse per cycle. For nonrectified x-ray
tube circuits, electrons may propagate from
the anode to the cathode if the anode reaches
a temperature high enough to release electrons
by thermionic emission. In addition, without
the rectifier circuit in place, only one-half of
each cycle of the applied voltage will produce
x rays.In a single-phase generator, the low-voltage
AC waveform is first transformed into high-volt-
age AC, rectified in the bridge rectifier circuit,
and then applied to the x-ray tube as a pulsat-
ing DC. For a significant fraction of time, the
applied voltage is much less than the peak volt-age (Fig 1 5). During the time of low voltage, ra-
diation output is reduced because of inefficient
x-ray production and space charge-limited op-
eration occurs. In this situation, the tube out-
put varies considerably, increasing and decreas-
ing in phase with the applied pulsating DCwaveform. At tube potentials above 40 kVp,
operation is emission limited, which explains
the flat peak of the milhiampere curve in Figure15. Actual applied voltage across the x-ray tube
electrodes does not vary as much as depicted
because of cable capacitance effects (stored
charge chiefly depends on the length of the
cables). Cable capacitance creates a lag in the
applied potential difference across the x-ray
tube that is related to the delay in discharge of
the cables (Fig 15).
120#{176}phase
�1�
Composite 3 phase
240#{176}phase
Transformer
60 Hz AC line voltage input for each of 3 phases phase map:
November-December 1997 Seibert U RadioGraphics U 1545
00 (360#{176})phase � I00
c;� �o#{176}180\�270#{176}�/’�360#{176}
240#{176}
Bridge
Rectifiers �
Output voltage:
6 pulses per cycle
Figure 16. Diagrams demonstrate the output voltage for a three-phase, full-wave-rectified
generator circuit. Diagram in the upper right illustrates three phases. The availability of linesources with 1 20#{176}different phase (upper left) can be combined to produce a rectified out-
put voltage with small variations in peak to minimum voltage (lower right). (Redrawn, with
permission. from reference 6.)
Voltage ripple describes the percentage dif-
ference between the minimum and maximumvoltages divided by the maximum voltage in
the circuit: % voltage ripple = [(Vm� _
V,uaxl 100. In theory, a single-phase generator
with either one- or two-pulse output has 100%
ripple. For a single-phase one-pulse generator
(which does not produce x rays during one-half of the cycle) to deliver the same output of
radiation intensity as produced by a full-wave-
rectified (two-pulse) generator, a longer expo-
sure time is required. The actual voltage ripple
for a single-phase two-pulse generator is actu-
ally less than 100%, due to cable capacitance
effects. Single-phase generators have the high-
est ripple, which results in suboptimal perfor-
mance of the x-ray tube, leading to slightly
higher patient dose and potentially poorer im-
age quality compared with those produced by
other generators that have less voltage ripple.
. Three-Phase GeneratorsThree-phase generators use the three separate
input lines supplied by the electric company.
These input lines are identical in frequency and
voltage amplitude but differ in the phase of the
AC waveforms. Phase is characterized as the
delay between the sinusoids with respect to a
reference starting point. If a single AC wave-
form starts at the origin (0#{176}),in the first half-
cycle, it wifi reach a maximum peak at 90#{176},
pass through 0 V at 180#{176},reverse polarity,
reach a negative peak at 270#{176},return to the ori-
gin at 360#{176},and repeat. Because the repetition
of the waveform occurs over an angular distri-
bution of 360#{176},two extra waveforms withphase differences of 1 20#{176}(ie, starting at I 20#{176}
and 240#{176})will symmetrically fill the phase map
(Fig 16). Three-phase power supplies consist of
three separate AC lines with relative phase dii-
ferences of 0#{176}, 1 20#{176}, and 360#{176}. When these
lines are simultaneously combined, the overall
voltage variation in the rectified output is sub-
stantially reduced. Separate transformation of
the individual line voltages, rectification of the
high-voltage output, and combination of the
output rectified waveform results in a very low
voltage ripple. Increased efficiency in x-ray
production and higher tube output for a given
tube current is achieved, because the average
applied voltage during the exposure is very
close to the peak voltage.
[Rectifiers I
1546 U Imaging & Therapeutic Technology Volume 17 Number 6
I�T�
“Delta” wiring confi
Figure 17. Circuit diagram of a three-phase six-pulse generator transformer and rectifier
circuit. Three primary and three secondary- transformers are wired together in a delta con-
figuration. Each transformer has its own line voltage source, and all three share a common
iron core. Current is routed through the bridge rectifier circuit to ensure correct I)OlafltY of
the cathode and anode within the x-ray tube for all three simultaneous line voltages. produc-
ing a total of six pulses per cycle.
A three-phase six-pulse generator is corn-
posed of a high-voltage circuit consisting of
two delta configurations of transformers, oneon the primary side and one on the secondary
side of the circuit (Fig 17). Each delta configu-
ration contains three separate yet intercon-
nected transformers, each operating indepen-
dently with line voltages of different phases. A
bridge rectifier circuit containing six diodes
(two diodes per transformer) is necessary to
ensure the correct polarity and electron flow
through the x-ray tube. The designation “six-
pulse” refers to the number of pulses per cycle
produced from the secondary delta configura-
tion. Output voltage for a six-pulse system var-
ies from the peak voltage on the order of 13%-
25% (compared with 100% variation for a
single-phase generator).
A more sophisticated three-phase 1 2-pulse
generator consists of an input delta configura-tion and two secondary groups of transformers
configured in a delta and a w�’e (Fig 18). In ef-
feet, this configuration represents two six-
pulse generators coupled to the x-ray tube that
are offset from each other by an additional 60#{176}
resulting from the wiring configuration and
phase variation produced by the wye-config-
ured transformers. The number of output
pulses per cycle is doubled, producing 12
pulses per cycle, with an applied voltage s-aria-
tion on the order of 5%- 10% of the peak volt-
age (Fig 19).
In three-phase generator designs, high-pow-
ered mode (or tetrode) circuits on the second-
ary side of the circuit are needed to control the
x-ray exposures. As explained, these switches
can turn the x-ray beam on and off during any
phase of the applied voltage within extremely
short times (millisecond or better accuracy).
Although their voltage ripple is very desirable
(3%-25%) and exposure control is fast and ac-
curate, three-phase generators require a sub-
stantial amount of hardware and electronics,
are more expensive than single-phase systems,
take longer to install, and occupy signifIcantly
more floor space.
. Constant Potential GeneratorsConstant potential generators are so named be-
cause of their ability to produce a high-voltage
waveform with essentially no voltage ripple
(essentially DC or “constant potential”). This is
the preferred waveform, because the electrons
are accelerated at an even potential difference
throughout the exposure. A higher effective
energy beam and greater tube output is ob-
tamed with a DC waveform.
Constant potential generators are even more
complex and bulky than three-phase systems.
A constant potential generator contains two
high-voltage electron tubes (triodes or tet-
AppliedTube
Voltage
resultant 12-pulse rectifiedwaveform across x-ray tube
November-December 1997 Seibert U RadioGrapbics U 1547
I Pdrnaiy
In�’�1 _
/Wye or Star
w,nng configuration
Figure 18. Circuit diagram of a
three-phase 12-pulse generator trans-
former and rectifler circuit. A see-
ond set of three secondary trans-
former windings is configured in a
s�5.e configuration and is connectedto a second bridge rectifier circuit.
The output of the delta and wye see-
ondary transformers is two six-pulse
rectified waveforms.
AnodeVoltage
CathodeVoltage
6-pulse voltage rectifiedwaveforms per electrode;note 60#{176}phase difference
between anode and cathodeFigure 19. l)iagram depicts the ef�
feet of the 12-pulse generator design
on the AC waveform. In the three-
phase 1 2-pulse generator. the two
rectifier circuits are symmetrically
placed on the anode and cathode. A
60#{176}phase shift in the waveforms is
produced by the specific delta and
wye configurations. Because the in-
dividual six-pulse outputs per recti-
fier circuit do not overlap, a total of1 2 pulses per cycle is produced (bot-
tom).
rodes) placed on the cathode and anode sides
of the secondary circuit. These tubes control
both exposure (switching the tube voltage Ofl
and off with about 20-�.tsec accuracy) and the
high voltage in the secondary circuit (with 20-
50-�.tsec adjustment capability). Voltage divid-
ers measure the actual kilovoltage, which is
compared with the reference kilovoltage in a
comparator circuit that controls the grid dcc-
trode (and thus the current and potential differ-
ence) of the triode or tetrode tubes. This
closed-loop feedback ensures extremely rapid
adjustment in kilovoltage and a nearly ripple-
free (DC) waveform. Fast response time of the
frilly electronic adjustment allows repetition
rates of 500 frames per second or more.
However, even though the constant poten-
tial generator produces the superior voltage
waveform and has the highest output power
compared with any other generator type, the
high hardware cost, high operational expense
(eg, calibrations), and immense space require-
ments have diminished its popularity for new
or replacement radiographic systems.
. High-Frequency Inverter Generators
The high-frequency inverter generator is the
current state-of-the-art choice for diagnostic ra-
diographic systems and has been available for
the past 10- 1 5 years, during which time signifi-
cant improvements have been made. Its name
is descriptive of its function: conversion of a
low-frequency, low-voltage input into a high-
frequency, low-voltage waveform that pro-
duces a high-frequency, high-voltage output
waveform.
Input voltage Rectifier Smooth
�1V’�\ /W\
IIOV 60Hz
Transformer
lnverter
“IA”
Rectifier
AC 500 - 40000 Hz
X-ray tubeSmooth
1548 U Imaging & Therapeutic Technology Volume 17 Number 6
Figure 20. Diagram illustrates the
steps by which a high-frequency gen-
erator produces a high-frequency,
high-voltage AC waveform. The AC
input voltage is converted to DC. An
inverter circuit chops the DC inputinto high-frequency AC of 500-
40,000 Hz (dependent on severalfactors explained in the text), which
is applied to the high-voltage trans-
former. High-voltage conversion,
rectification, and smoothing provide
a potential difference to the x-ray
tube with minimum voltage ripple.
(Redrawn, with permission, from
reference 6.)
Either single- or three-phase input lines can
be used to produce an output voltage with
minimal ripple characteristics (the three-phase
source provides greater overall power). The
AC input power is first converted into a low-
voltage DC waveform by rectification and
smoothing. An inverter circuit “chops” the DC
into a high-frequency AC square wave, which
is input into the high-voltage transformer to
produce a high-voltage, high-frequency AC
waveform. Subsequent rectification and capaci-
tance smoothing produce a potential difference
across the x-ray tube with very small voltage
ripple (Fig 20). In function, the generator fre-
quency is variable, since it depends on the set
tube potential, the tube current, and the fre-
quency-to-voltage characteristics of the trans-
former.
In addition to the components just de-
scribed, a high-frequency generator contains
high-voltage capacitors in the secondary circuit
that produce a voltage across the x-ray tube de-
pendent on the accumulated charge within
each capacitor (Fig 2 1). The voltage-charge-ca-
pacitance relationship V = QIC describes this
process, where V = the voltage across the ca-
pacitor, Q = the charge on the capacitor, and C
= the capacitance in farads. The generator op-
crates with a closed-loop regulation circuit,whereby during an exposure, feedback circuits
monitor both the tube voltage and tube cur-
rent. Instantaneous adjustments are made to
correct for variations in kilovoltage and mill-amperes that occur during the exposure.
In the case of kilovoltage, a voltage corn-
parator or oscillator measures the difference
between the reference voltage (a calibrated
value proportional to the requested kilovolt-
age) and the actual kilovoltage measured
across the x-ray tube by a voltage divider (kilo-
volt sense circuit). Trigger pulses generated by
the comparator circuit are produced with a fre-
quency proportional to the voltage difference:
A large discrepancy between the compared
voltages produces a high trigger pulse fre-
quency, whereas no difference produces few
or no trigger pulses. For each trigger pulse, the
DC-AC inverter circuit produces an output AC
waveform, which is converted into a high-volt-
age output pulse by the transformer. Two high-
voltage capacitors symmetrically balanced on
the cathode and anode of the x-ray tube store
the charge and create a potential difference. If
the actual voltage measured by the kilovolt
sense circuit is low, the trigger pulse rate to
the inverter increases, actuating the high-volt-
age transformer, the output of which ‘dumps”charge into the high-voltage capacitors to raise
kilovoltage across the x-ray tube. When the x-
ray tube potential reaches the desired value,
the output pulse rate of the comparator circuit
settles down to an approximately constant
value and recharges the high-voltage capacitors
only when the actual tube voltage drops below
a predetermined limit.
Tube current is regulated in a manner anal-
gous to that used for kilovoltage, with a resis-
tor circuit sensing the actual miffiamperes (mil-
liampere sense voltage) and comparing that
value to a reference milliampere value (milli-
ampere reference voltage). If the milliampere
sense value is too low, the voltage comparator
increases the trigger pulse frequency, which
boosts the power to the filament to raise tern-
perature and increase thermionic emission of
electrons. The closed-loop feedback circuit
eliminates the need for space charge compen-
sation and adjusts for filament aging effects.
The design of the high-frequency inverter
generator has several advantages. Because the
input line voltage is initially converted from anAC to a DC waveform, either single- or three-
+kV/2
Typical voltage
waveform
TVThvr\
kV
ripple
100%
100%
_\ 13%-25%
� 3% . 10%
ence 6.)
November-December 1997 Seibert U RadioGraphics U 1549
Figure 21. Block diagram shows the internal components and feed-hack circuits of a general-purpose high-frequency generator. Voltage
and current are maintained by comparing the actual kilovolt and milli-
ampere values (directly measured during the exposure) to referencevalues selected at the control console. High-voltage capacitors on the
secondary circuit supply the potential difference across the x-ray tube
and are continuously charged as needed to maintain the desired volt-
age. Autotransformers and space charge compensation circuits are not
required because of the closed-loop feedback circuits maintained by
this generator.
Generatortype
Single-phase 1-pulse(self rectified)
Single-phase 2-pulse(full wave rectified)
3-phase6-pulse
3-phase12-pulse
Medium-high frequencyinverter
ConstantPotential
Figure 22. Chart illustrates the4% . 15% voltage waveforms and voltage
ripple percentages for the x-ray gen-
erators described in this article. (Re-<2% drawn, with permission, from refer-
phase input lines can be used. Closed-loop
regulation circuits, which ensure reproducible
and accurate kilovoltage and tube current, pro-
vide extremely accurate exposures and obviate
autotransformers. Output voltage ripple of
high-frequency inverter generators is kilovolt-
age dependent and is similar to that of three-
phase 1 2-pulse generators (Fig 22). Transform-
ers operating at high frequencies are more effi-
cient, more compact, and less costly to manu-
facture. Finally, most high-frequency genera-
tors have an excellent service record and are
relatively easy to repair.
Currently, the high-frequency inverter gen-
erator is the most desired type for all but a few
selected applications (eg, extremely high
power needs or extremely fast kilovoltage
switching or short [< 1 msec] exposure times).
In only rare instances is the constant potential
generator possibly a better choice.
1550 U Imaging & Therapeutic Technology Volume 17 Number 6
Figure 23. Diagram of a common mammographic unit illustrates the components of the
phototimer system. Modern phototimer systems in mammography have microprocessor
controls to adjust and compensate for unusual variations in the acquisition process. (Re.
drawn. with permission. from reference 6.)
U METhODS OF EXPOSURE CONTROL
Most single-phase generators and low-power x-
ray systems use simple mechanical switches
coupled to either mechanical or electronic tim-
ers to open and close the circuit on the pri-
mary side of the high-voltage transformer. For
these applications, the minimum exposure
time is 1/1 20-second (8-msec), since the
switch, once closed, stays closed as long as
power is applied. During each half-cycle, the
voltage passes through zero with the power in
the circuit also at zero, which allows easy re-
lease of the mechanical switch. Thus, the cx-
posure times allowed occur in 1/1 20-second in-
crements.
For three-phase and constant potential gen-
erators, high-voltage triode, tetrode, or pentode
switches initiate and stop the exposure on the
secondary side of the circuit under the control
of an electronic timer (or phototimer feed-
back) that is connected to the grid electrode.
Small negative voltage biasing on the grid (eg,
1 ,000 V) produces an electric field that over-
whelms the large potential difference (eg,
100,000 V) because the grid is close to the
cathode of the thyratron (gas-filled triode) or
thyristor (solid-state triode) switch. The flow of
electrons is stopped when the grid is activated.
Because the switch can he opened and closed
independent of the power on the circuit, expo-
sure times of 1 rnsec and less are possible.
The high-frequency invei-ter generator typi-
cally uses electronic switching on the priniat-y
side of the high-voltage transformer to initiate
and stop the exposure. The high-frequency
waveform characteristics of the generator cir-
cuit enable relatively rapid switching. which al-
lows exposure times typically a� short as 2
msec.
. PhototimersPhototimers are implemented in lieu of manual
exposure settings. These devices measure the
actual exposure incident on the x-ray receptor
(eg, screen-film detector) and turn off the x
rays when the proper optical density is oh-
tamed. An accurately adjusted phototimer pro-
vides consistent film optical density and corn-
pensates for variations in thickness and attenu-
ation within the patient.
The phototimer system consists of a radia-
tion detector, amplifier, density selector, corn-
parator circuit, termination switch, and backup
tinier. X rays transmitted through the patient
instantaneously generate a small signal in an
ionization chamber (or other radiation-sensitive
device such as a solid-state detector). An ampli-
tier boosts the signal, which is fed to a voltage
comparator circuit and integrated. When the
accumulated signal equals a preselected refer-
ence value, a pulse is generated to terminate
Cassette stand Film-screen cassette
:: � � � � :� � � � . � #{149}
.�.1. 1.4�3 2 1 0+1+2+3+4
I__I__I I I I�;:� � � � � �
-�.‘‘�-
Density control setting
AEC detector locations
November-December 1997 Seibert U RadioGrapbics U 1551
Figure 24. Diagram demonstrates the typical location of automatic exposure control
(AJIC) detectors in a conventional projection radiographic system. Usually, three ionization
detectors are placed in front of the screen-film detector in the positions shown. Signals from
either or all of the detectors can be selected by the automatic exposure control electronics
circuit. This arrangement allows for global or regional compensation for variations in ana-
tomic structure. which is particularly important for chest imaging. (Redrawn, with permis-
SiOfl. from reference 6.)
the exposure. A radiographic system with a
properly adjusted phototirner circuit will pro-
duce films with the desired optical density.
The reference voltage can be adjusted with a
user-selectable density selector switch on the
generator control panel to modify the optical
density in steps of±lO%-20% from the normal
density position.
In a marnmographic system, the ionization
chamber of the phototimer is positioned be-
hind the screen-film cassette (Fig 23). This
placement is necessary to avoid producing arti-
facts from the chamber on the film. For most
general diagnostic imaging, the phototimer sen-
sors are placed in front of the detector (Fig 24)
because the high kilovolt spectrum used
makes the chamber transparent to x rays.
A backup timer ensures that the x-ray expo-
sure is terminated after a predetermined time
in the event of a phototimer detector or circuit
failure. Often in mammographic examinations,
the backup timer will be activated because the
operator inappropriately selected a low tube
potential that resulted in x rays which failed to
penetrate the thick, dense breast.
. Automatic Exposure ControlsAutomatic exposure control is another name
for a l)liototimer, and its circuitrr is an integral
part of most modern x-ray generators. Its cir-
cuitry is composed of detectors that measure
the exposure incident OIl the x-ray receptor
and that provide feedback signals to the timing
switch of the generator. Predetei-niined sensi-
tivity settings are matched to the speed of the
x-ray receptor to achieve optimal-quality radio-
graphs by turning off the power to the x-ray
tube after the necessary x-ray exposure.
To compensate for different x-ray penetra-
tion through distinctly different anatomic struc-
tures, three sensors are typically used in a
chest or table cassette stand with automatic cx-
posure control capability (Fig 24). The tech-
nologist can use the exposure signal from one,
two, or three of the sensors either indepen-
dently or together, depending on the particular
application. For instance, in chest radiography,
the outside sensors are often the only detectors
activated so that the x-ray beam transmitted
through the lung areas determines the expo-
sure time. In this way, “burn-out” of the lungs
that might occur if the transmitted x-ray flux
were measured solely under the highly attenu-
ating mediastinum is prevented.
In fluoroscopic systems, automatic bright-
ness control circuitry, which is similar to auto-
matic exposure control circuitry, is used to
provide optimal kilovolt and milliampere set-
tings for dynamic image display. When auto-
matic brightness control is enabled, the overall
image brightness on the display monitor re-
mains consistent because the generator pro-
�‘ides changes in the fluoroscopic parameters
to compensate for changes in the thickness
and composition of the patient that occur with
movement during the procedure.
500 -
Example: 50 mAs necessary for exposure:mAs delivered a area under curve
mAs is equal in both situations
400 -
E
300
200
With falling load generatorMaximum variable mA x 0.08 sec
Without falling load generator:400 mA x 0.125 sec
Falling load curve
I ‘I I � � � � � I I I
0.1 0.2 0.3 0.4 0.5 0.6 0.7
Exposure time (sec)
1552 U Imaging & Therapeutic Technology Volume 17 Number 6
Figure 25. Plot demonstrates the advantage of using a falling load
generator. In the example plotted, a particular examination to be per-formed at a specific tube potential requires 50 mAs to record the
proper optical density on film, and a maximum constant tube current
of 400 mA (as determined with a tube rating chart) can be tolerated by
the focal spot. For these examination parameters, an exposure time of1 25 msec is required. With a falling load generator, however, it is de-
termined that a maximum tube current of 600 mA can be delivered, if
applied for a short duration, much less than the total exposure time.Even though the tube current continuously falls during the exposureaccording to a predetermined power load curve, a total of 50 mAs
equivalent output can be completed within an exposure time of 80
msec, or about 45 msec less than the scenario with a constant maxi-
mum tube current. The shaded bars have the same area; therefore, thesame film optical density is achieved with both techniques. (Redrawn,with permission, from reference 6.)
. Faffing Load GeneratorsThe falling load generator works with the pho-
totimer or automatic exposure control systems
to acquire an image in as short an exposure
time possible. This type of generator is pro-
grammed to deliver the maximum possible
tube current for the selected tube potential;
this maximum value is determined from the fo-
cal spot selected and the instantaneous heat
load characteristics of the x-ray tube anode.
The preprogrammed circuitry of the generator
determines the maximum power load limit of
the x-ray anode and then continuously reduces
the power as the exposure continues so that
the heat load limit of the anode is not cx-
ceeded during the exposure time (Fig 25).
Thus, use of a falling load generator allows the
optimal film density to be recorded in the
shortest possible exposure time. In addition,
shorter exposure times may help decrease im-
age blurring caused by involuntary or voluntary
patient motion.
U SELECTION OF GENERATOR POWER
AND OPTIONS
. Power Rating: Generators and Focal
Spots
The power rating of generators is determined
by the characteristics of the input voltage, the
filament transformer, and the high-voltage
transformer. The typical unit for power is the
Table 1
Power Ratings of X-ray Generators and Focal Spot Sizes
Focal Spot
Application Generator Power Rating (kW) Size (mm) Power Rating (kW)
Dental, mammography � 10 0.1-0.3 1-10
Fluoroscopy, general � 50 0.3-0.6 10-30
Angiography, interven- � 150 0.8-1.0 50-80
tional
Contant potential � 200 1 .2- 1 .5 80-125
November-December 1997 Scibert U RadioGraphics U 1553
kilowatt, which is the product of the voltage
and tube current expressed in kilovolts and
milliamperes, respectively. Strictly speaking,
the voltage is assumed to be a pure DC wave-
form with 0% ripple. For voltage waveforms
with substantial ripple, the power rating is a
product of the effective voltage (the root-mean-
square voltage [ V,.,,,,]) and the effective current.
The generator power capability is typically
benchmarked at an exposure time of 0. 1 see-
ond for the maximum milliamperage and
kilovoltage. From this definition, a 100-kW gen-
erator can therefore provide 1 ,000 mA at 100
kVp for a 0. 1-second exposure. Lower or
higher tube potentials can support higher or
lower tube current output, respectively, within
a range of 20-30 kV. For example, when the
same 100-kW generator is operated for a 0.1-
second exposure at 80 kVp, the tube current
will be about 1,200 mA; at 120 kVp, the maxi-
mum tube current will be about 800 mA.
For single-phase generators with substantial
voltage ripple, the power rating is reduced by
about 30% because of the lower effective tube
potential caused by the large variation in the
voltage. A generator operating at 100 kYp with
a single-phase waveform that is rated at 1,000
mA for 0. 1 second has a power rating of about
70 kW.
The power output of the generator must
also be matched with the power dissipation
(loadability) characteristics of the focal spot
and anode for a specific x-ray tube. A bigger fo-cal spot will have a larger power rating than a
small focal spot. Power loading of the x-ray
tube can be determined from the single expo-
sure rating chart in a manner analogous to that
described for the generator power rating.Power ratings of generators and focal spots
are specified in Table 1 . Generator power re-
quirements should be considered in terms of
the imaging application. In addition, it makes
no sense to have a very high rated generator
and then be limited by the focal spot power
rating. The appropriate equipment (x-ray gen-
erator and x-ray tube) must be specified for the
intended imaging tasks, so that efficiency of
economy, size, and utilization are maximized.
. Tube Rating Charts
Tube rating charts provide guidance regardingthe maximum allowable deposition of powerfor a single exposure, multiple exposures, and
continuous x-ray exposure for a specific x-ray
tube and x-ray generator combination. The
power limits are determined by the selected
tube potential, tube current, exposure time, fo-
cal spot size, anode rotation speed, and genera-tor type. Historically, the heat unit has been
used to indicate the power deposited on the
anode for single exposures. It is calculated as
the product of the tube potential, tube current,
and exposure time in seconds. To determinethe continuous power deposition during fluo-
roscopic procedures, the heat unit per second
is calculated as the product of tube potential
and tube current (which follows from the defi-
nition of the heat unit).
Table 2Generator Waveform Characteristics andAverage Voltage
1556 U Imaging & Therapeutic Technology Volume 17 Number 6
Figure 28. Charts of anode and housing thermal characteristics are illustrated. Anode heat input
and heat dissipation rates are specific to a given x-ray tube. In the left chart, the largest y-axis value
most always indicates the maximum anode heat load limit (measured in kilo heat units). Heat dissipa-
tion is characterized by the exponentially decreasing curve, and heat input rate (heat units per see-
ond) is usually shown as a family of individual curves to determine the amount of heat accumulating
as a result of fluoroscopic exposures. The right chart illustrates how use of a heat exchanger systemcan accelerate the cooling rate of the housing. (Redrawn, with permission, from reference 6.)
In modern x-ray generators, system logic cir-
cuits prohibit any combination of tube poten-
tial, tube current, and exposure time that
would exceed the tube rating capabilities of
the x-ray tube and the chosen focal spot. Forangiography, however, because multiple expo-
sure studies are common, one must consult the
multiple exposure rating chart (containing a se-
ties of tables) to determine the ability of the
system to perform the requested examination
series without damaging the x-ray tube.
. Thermal Characteristic Charts for
X-ray Tube Anodes and HousingsHeat loading and dissipation rates for the x-ray
tube anode and x-ray tube housing must be
considered for extended imaging examinations
(eg, combined fluoroscopy and radiographicimaging for interventional angiography). The
anode and housing thermal characteristiccharts are used for this purpose (Fig 28).
In the anode chart, the x axis indicates time
in minutes, and the y axis indicates the heat
units accumulated in the anode as a result of
previous exposures or continuous fluoroscopic
input. The largest value on the y axis indicates
the maximum power load rating for the x-ray
tube. In the anode chart example, this value is
140,000 heat units. Single and multiple expo-
sures instantaneously deposit energy on the an-
ode. For a three-phase generator, the heat unit
value must be multiplied by 1 .35- 1 .40 to deter-
mine the actual number of heat units depos-
ited. During continuous fluoroscopy, the
amount of heat deposited on the anode is de-
termined by the heat units per second input.Several heat input curves corresponding to
various values of heat units per second (ie,
tube potential times tube current) are used to
estimate the total amount of heat deposited on
the anode. Many of these curves at low miii-
amperes reach a maximum when the heat in-
put rate equals the heat dissipation rate. For
situations in which the exact heat units per
second input curve is not available, interpola
This article meets the criteria for 1. 0 credIt hour in Category 1 of the AMA Physician ‘s Recognition
Award. To obtain credit, see the questionnaire on pp 1527-1532.
November-December 1997 Seibert U RadioGrapbks U 1557
tion between the two curves above and below
the desired heat units per second value is nec-
essary to estimate the accumulated heat after a
specified fluoroscopic exposure time.
The anode-cooling curve describes the rate
at which heat is reemitted from the anode. A
much steeper slope is evident at high heat
loads, indicating that much faster cooling oc-
curs with a hotter anode. For example, a tube
with 140,000 heat units on the anode will lose
50,000 heat units (to 90,000 heat units) in
about 2 minutes, whereas a tube with 90,000
heat units on the anode will lose 50,000 heat
units (to 40,000 heat units) in approximately
3’/2 minutes. The anode thermal characteristics
chart is used with single and multiple exposure
charts to determine whether a single or series
of subsequent exposures are permitted based
on the amount of heat already accumulated.
In the tube housing chart, a maximum heat
accumulation is indicated on the y axis. In this
situation, thermal expansion of the insulation
oil surrounding the x-ray tube insert occurs. Ex-
ceeding the maximum allowed level can causecatastrophic failure of the housing or x-ray
tube. Like the anode, the housing cools fasterwhen it is hotter. In some x-ray tubes, heat cx-
changers are available to cool the housing and
its components more rapidly.
Specific situations in which the anode ther-
mal characteristic chart would be used include
(a) determination of the maximum anode heat
loading value, (b) calculation of the amount of
heat deposited on the anode as a result of fluo-
roscopic and radiographic techniques, and
(c) determination of the length of time required
for cooling before initiating another imaging se-
quence on an already hot anode. For extended
use of the x-ray system, the housing thermal
characteristics chart would also be used in situ-
ation C.
U SUMMARY
The x-ray generator provides the power neces-sary to produce x rays within the x-ray tube.