X-Rav Source

Cross-Reference to Related Applications

This application is a continuation-in-part of two patent applications: (1) Serial No. 08/375,501, filed January 17, 1995, which is a continuation of U.S. Patent Application Serial No. 08/042,742, filed April 5, 1993, now abandoned; and (2) International Patent Application Serial No. PCT/US94/03737, filed April 5, 1994.

Background of the Invention

1. Field of the Invention The present invention pertains to the field of charged particle beam generators and x-ray tubes, and more specif¬ ically, scanning beam x-ray sources.

2. Description of Related Art

Real-time x-ray imaging is increasingly being required by medical procedures as therapeutic technologies advance. For example, many electro-physiologic cardiac procedures, peripheral vascular procedures, PTCA procedures (per¬ cutaneous transluminal catheter angioplasty) , urological procedures, and orthopedic procedures rely on real-time x-ray imaging. In addition, modern medical procedures often require the use of instruments, such as catheters, that are inserted into the human body. These medical procedures often require the ability to discern the exact location of instruments that are inserted within the human body, often in conjunction with an accurate image of the surrounding body through the use of x-ray imaging.

A number of real-time x-ray imaging systems are known. These include fluoroscope-based systems where x-rays are projected into an object to be x-rayed and shadows caused by relatively x-ray opaque matter within the object are displayed on the fluoroscope located on the opposite side

of the object from the x-ray source. Scanning x-ray tubes have been known in conjunction with the fluoroscopy art since at least the early 1950s. Moon, Amplifying and Intensifying the Fluoroscopic Image bv Means of a Scanning X-raγ Tube, Science, October 6, 1950, pp. 389-395.

Reverse-geometry scanning beam x-ray imaging systems are also known. In such systems, an x-ray tube is em¬ ployed to generate x-ray radiation. Within the x-ray tube, an electron beam is generated and focussed upon a small spot on the relatively large anode (transmission target) of the tube, inducing x-ray radiation emission from that spot. The electron beam is deflected (electro- agnetically or electrostatically) in a raster scan pattern over the anode. A small x-ray detector is placed at a distance from the anode of the x-ray tube. The detector typically converts x-rays which strike it into an electrical signal in proportion to the detected x-ray flux. When an object is placed between the x-ray tube and the detector, x-rays are attenuated by the object in proportion to the x-ray density of the object. While the x-ray tube is in the scanning mode, the signal from the detector is inversely proportional to the x-ray density of the object.

Examples of known reverse-geometry scanning beam x-ray systems include those described in United States Patent No. 3,949,229 to Albert; United States Patent No. 4,032,787 to Albert; United States Patent No. 4,057,745 to Albert; United States Patent No. 4,144,457 to Albert; United States Patent No. 4,149,076 to Albert; United States Patent No. 4,196,351 to Albert; United States Patent No. 4,259,582 to Albert; United States Patent No. 4,259,583 to Albert; United States Patent No. 4,288,697 to Albert; United States Patent No. 4,321,473 to Albert; United States Patent No. 4,323,779 to Albert; United States Patent No. 4,465,540 to Albert; United States Patent No. 4,519,092 to Albert; and United States Patent No. 4,730,350 to Albert.

In a typical known embodiment of a reverse-geometry scanning beam system, an output signal from the detector is applied to the z-axis (luminance) input of a video monitor. This signal modulates the brightness of the viewing screen. The x and y inputs to the video monitor are typically derived from the signal that effects deflec¬ tion of the electron beam of the x-ray tube. Therefore, the luminance of a point on the viewing screen is inverse¬ ly proportional to the absorption of x-rays passing from the source, through the object, to the detector.

Accordingly there is a need for a source of x-rays that is both safe and economical as well as capable of being able to be positioned quickly and accurately across the face of a target anode.

Summary of the Invention

The x-ray source of the present invention comprises a charged particle beam generator and an anode assembly. The preferred charged particle beam generator is an electron beam source comprising a single direct electri- cal connection for providing voltage to the electron gun. The power for the active circuits in the high voltage terminal of the electron beam source is provided by a unique isolation transformer that has minimal loss¬ es and generates controlled magnetic flux. The generated electron beam is controlled through a series of dynamic and static focus coils and moved across the face of the target anode by a stepping coil assembly comprising x and y deflection coils as well as an x step and preferably a y step coil. Further, to minimize power usage, a control grid pinches off the electron beam during the stepping of the beam.

The entire x-ray source is packaged in a small form factor with sufficient safety features and that will allow for mounting of the source in traditional C-arms for use in medical applications without fear of danger to the patient or the treating physician.

These and many other objects and advantages of the present invention will become apparent to those of ordi¬ nary skill in the art from a consideration of the draw¬ ings and the description of the invention contained herein. The principles of the present invention may be employed in any application, medical or industrial.

Brief Description of the Drawings

Fig. 1 is a diagram showing the basic components of a preferred scanning beam x-ray imaging system. Fig. 2 is a cross-sectional side view of a preferred scanning beam x-ray tube.

Fig. 3 is an enlarged partial cross sectional repre¬ sentation of a preferred target and collimation grid. Fig. 4 is an exploded perspective view of a preferred high-voltage terminal assembly.

Fig. 5 is a perspective view of a preferred circuit board assembly.

Fig. 6 is a perspective view of a preferred isola¬ tion transformer. Fig. 7 is a diagram of a representative inner struc¬ ture of a preferred isolation transformer secondary assembly.

Fig. 8 is a cross-sectional view of a preferred isolation transformer primary assembly. Fig. 9 is a diagram of a representative inner struc¬ ture of an alternately preferred isolation transformer primary assembly.

Fig. 10 is a diagrammatic representation of the magnetic field lines between the primary and secondary of the preferred isolation transformer.

Fig. 11 is a diagram of a preferred fast deflection yoke structure.

Fig. 12 is an enlarged perspective diagram of the preferred deflection yoke and focus coil alignment structures.

Fig. 13 is an exploded view of the deflection yoke and focus coil alignment structures of Fig. 14.

Fig. 14 is a diagram of a preferred x-ray tube scan¬ ning an electron beam in a raster scan pattern. Fig. 14A-F are graphical representations of the current applied to deflection coils to move an electron beam in a raster scam pattern.

Fig. 15 is a diagram of a preferred scanning x-ray tube scanning an electron beam in a serpentine pattern. Fig. 15A-G are graphical representations of the current applied to deflection coils to move an electron beam in serpentine pattern.

Description of the Preferred Embodiments

Turning to Fig. 1, an embodiment of a presently preferred x-ray source employed in a reverse geometry scanning beam x-ray imaging system is diagrammed. The x-ray source 10 preferably comprises an x-ray tube and a high voltage electron beam source. The high voltage electron beam source is preferably connected to an ad- justable high-voltage power supply capable of generating approximately -70kV to -120kV. At this voltage level, scanning x-ray source 10 produces a spectrum of x-rays ranging to 120 keV. Scanning x-ray source 10 includes deflection coils 20 under the control of a scan genera- tor 30. An electron beam 40 generated within high-volt¬ age terminal 803 is scanned across a grounded anode target 50 in a predetermined pattern. For example, the predetermined pattern may be a raster scan pattern, a serpentine (or "S " shaped) pattern, a spiral pattern, a random pattern, a gaussian distribution pattern centered on a predetermined point of the target anode, or such other pattern as may be useful to the task at hand. Presently preferred is the serpentine (or "S" shaped) pattern which eliminates the need in a raster scan pat- tern for horizontal "fly back."

As electron beam 40 strikes anode target 50 at focal spot 60, x-rays 70 are emitted in all directions. For simplicity, only a portion of the x-rays are shown. The x-rays preferably pass through a collimator toward the object 80 to be investigated. To optimize system per¬ formance of the presently preferred embodiment, a cone of x-ray photons should be generated that will diverge in a manner that will just cover the multi-detector array 110. This is preferably accomplished by placing a collimating element between the anode target 50 of the x-ray source 10 and the multi-detector array 110 and more preferably between object 80 and x-ray source 10. A more detailed explanation of the system parameters can be found in copending U.S. Patent Application Serial No. 08/386861.

The presently preferred configuration for this col¬ limating element is a grid of x-ray transmissive cylin¬ ders or apertures 140. Collimation grid 90 is designed to permit passage to only those x-rays whose axes are in a path that directly intersects the multi-detector array 110. Collimation grid 90 preferably does not move with respect to multi-detector array 110 while the system is in operation. Thus, as electron beam 40 is scanned across anode target 50, at any given moment there is only a single x-ray pencil beam 100 which passes through object 80 to multi-detector array 110.

The output of multi-detector array 110 is processed and displayed by control electronics/monitor 34 as an intensity value on a display monitor as described in copending and incorporated U.S. Patent Application Seri¬ al No. 08/386861.

Fig. 2 is a cross-sectional diagram of the presently preferred scanning beam x-ray source 10 which comprises an electron beam source 112 and a vacuum envelope assem- bly 176.

Electron beam source 112 is comprised of two alumi¬ num flanged cylinders 114 and 116 bolted to central

aluminum cylinder 118. Rear endplate 120, fabricated from aluminum with two sealed openings 134 and 136, is bolted to the rear of aluminum flanged cylinder 116. Front endplate 138, fabricated of aluminum with a sealed central apertured ceramic disc 128, is bolted to the front of aluminum flanged cylinders and the central cylinder. This method of construction permits electron beam source 112 to contain an insulating fluid within its confines, with ceramic disc 128 forming a seal be- tween electron beam source 112 and vacuum envelope as¬ sembly 176. It is to be understood that any suitably designed housing is within the contemplation of the current inventions.

High-voltage cable 122, extending from an external high-voltage power source (not shown) , supplies a poten¬ tial preferably variable between -70kV to -120kV to generate an electron beam. The energy of this electron beam is between 70kev and 120keV which corresponds to the potential applied through high voltage cable 122. The preferred method of connecting high voltage cable

122 to the high voltage terminal assembly 803 is through use of molded epoxy cable receptacle 142 which has an integral metal mounting flange. Receptacle 142 passes through opening 134 and is sealed to end plate 120 with an O-ring seal. High voltage cable 122 is fitted into a strain relief sleeve 156 fastened by screws through integral flange 158 to the integral metal flange of cable receptacle 142. A rubber end piece 162, prefera¬ bly ethylene propylene rubber, is shaped to conform with the conical orifice in cable receptacle 142 and is mold¬ ed directly to the end of high voltage cable 122. For assembly, rubber end piece 162 is coated with silicone grease and is tightly compressed into the orifice in cable receptacle 142 to minimize electrical breakdown along the interface between rubber end piece 162 and cable receptacle 142. High voltage cable 122 contains electrostatic shielding (not shown) which is connected

to ground within sleeve 156. Electrical contact is established between high voltage cable 122 and contact plate 146 by a conductive rod, thereby forming an elec¬ trical connection through conducting spring 152 to high voltage terminal assembly 803. Conducting spring 152 is preferably received in an indentation in the high voltage terminal endplate 154 of high voltage terminal assembly 803.

An insulating medium preferably surrounds high volt- age terminal assembly 803 to allow small distances be¬ tween the high voltage terminal assembly 803 and the outer walls of the electron beam source 112. Prefera¬ bly, the insulating medium should be able to allow a high electrical potential of at least -120kV to be im- pressed across this distance and maintained without electrical breakdown. The presently preferred insulat¬ ing medium is sulphur hexafluoride gas (SF ₆ ) , which is preferably maintained at a pressure of approximately 60psig and at a temperature less than 60°C. Other insu- lating media, such as transformer oil, can also be em¬ ployed in place of SF ₆ .

Preferably fitted within aperture 136 of rear endplate 120 is a feedthrough assembly 164 through which eight fiber optic cables enter electron beam source 112. For purposes of illustration only, the eight fiber optic cables are shown as a single cable 168 in Fig. 2. The fiber optic cables 168 are preferably sealed into feed¬ through assembly 164 by embedding them in epoxy resin in order to prevent leakage of the SF ₆ gas. High voltage terminal assembly 803 is preferably insulated to withstand the applied high voltage by means of ceramic disc 128 set within the front plate 138. High voltage terminal assembly 803 is also preferably mechanically supported by means of ceramic disc 128 only, to form a cantilever.

An isolation transformer 744 supplies power to the components within the high-voltage terminal assembly

803. The secondary 1271 of the isolation transformer 744 is located within the high-voltage terminal 803. The primary 1270 is disposed coaxially around the sec¬ ondary 1271, but is physically separated from the sec- ondary 1271 by an insulating gap filled with the SF ₆ insulating medium. The isolation transformer 744 is more fully discussed in the detailed descriptions of Figs. 6-10.

Vacuum envelope assembly 176, which is preferably at ground potential, generally comprises the entire struc¬ ture depicted in Fig. 2 to the right of the front end¬ plate 138. The interior of the vacuum envelope assembly 176 forms the pathway for the electron beam 40 from the high voltage terminal assembly 803 to the anode target 50. A tapered cylinder ring 262 extending from the front endplate 138 of the electron beam source 112 is welded to disc 264. An accelerating anode 184 with an axial through-hole is preferably screwed to the center of disc 264. Hereinafter, anode target 50 and acceler- ating anode 184 are referred to as target 50 and anode 184, respectively, for simplicity and clarity.

The interior of the vacuum envelope assembly 176 is maintained at a reduced pressure, preferably less than 10"mm Hg. Vacuum envelope assembly 176 is initially evacuated by means of a negative pressure source mounted on a vacuum stand attached to tube and flange assembly 183. During initial evacuation, vacuum envelope assem¬ bly 176 is preferably baked out at an elevated tempera¬ ture (>200°C) to outgas all items on the interior. During this high temperature bake-out, all components of electron beam source 112, except the front end plate 138, are preferably removed from x-ray source 10 so that they are not damaged by the high temperature. After bake-out, x-ray source 10 is re-assembled and con- ditioned, or high voltage processed, by operating the x- ray source at greater than normal voltage and current. The vacuum envelope assembly 126 is sealed off from the

vacuum stand by sealing the tube of assembly 183 using a conventional pinch-off tool. Thereafter the reduced pressure in vacuum envelope assembly 176 is preferably maintained through the use of getter-ion pump 182. Alternatively, vacuum envelope assembly 176 can be a "sealed" tube design which consequently eliminates the need for a getter-ion pump.

Electron gun 198 protrudes from the high-voltage terminal 803 through ceramic disc 128 into the vacuum envelope assembly 176. Electrode 126 preferably extends from the ceramic disc 128, surrounding the emitting end of electron gun 198. Electrode 126 and anode 184 are shaped to control the electrostatic field configuration in the accelerating space between electrode 126 and anode 184, thereby ensuring that electron beam 40 is correctly focussed through the axial hole in anode 184. Additional shaping of electrode 126 controls the elec¬ trostatic field configuration across the surface of ceramic disc 128 so that the chance of electrical break- down across the surface of disc 128 is minimized. On reaching anode 184, the electron beam 40 has acquired an energy expressed in electron volts substantially equal numerically to the voltage applied between electron gun 198 and anode 184. In its continuing path to target 50, electron beam 40 is preferably not subjected to any additional axial forces so upon impact at focal spot 60, the energy of electron beam 40 is essentially the same as that acquired at anode 184.

After leaving the axial hole of anode 184, electron beam 40 passes through a magnetic focus lens assembly 186 which is preferably a thin lens design comprising a cylindrical steel magnetic circuit with a U-shaped sec¬ tion. Static focus coil 185 is preferably wound on coil form 272 within this magnetic circuit. ^' Dynamic focus coil 187 is preferably located within the magnetic cir¬ cuit air gap and is preferably wound on a bobbin shaped coil form 270. Dynamic focus coil 187 is preferably

wound with substantially fewer turns of wire than static focus coil 184 so that dynamic focus coil 187 has a low inductance, thus permitting the current flowing in the dynamic focus coil 187 to be changed rapidly. Currents flowing in static focus coil 185 and dynamic focus coil 187 cause electron beam 40 to be brought to a focus at focal spot 60. When used with a collimation grid, the size of the focal spot 60 is important. It should be small enough to maximize the transmission of x-ray flux through the apertures in collimation grid 90 but, if it is too small, the resulting excessively high power den¬ sity concentrated in focal spot 60 could cause local melting of the surface of target 50. It has been found that a focal spot size of 0.3 mm is preferred when x-ray source 10 is used in conjunction with the collimation grid disclosed in U.S. Patent Application Serial No. 08/386861.

From the magnetic focus lens assembly, the path of electron beam 40 is preferably controlled by a two-part magnetic deflection system comprising slow deflection yoke 190 and fast deflection yoke 188 disposed coaxially around ceramic cylinder 180. The deflection yokes are described more fully in connection with the detailed descriptions of Figs. 11-15. Ceramic cylinder 180 is preferably formed of a ceramic material, as opposed to metal, because the rapidly changing magnetic fields produced by the deflection yokes, 190 and 188, would induce eddy currents in a metal cylinder which would inhibit penetration of the magnetic fields and so inter- fere with the accurate deflection of electron beam 40. Ceramic cylinder 180 is preferably formed of alumina, coated on the inside with a thin high-resistance coating of a nickel-chromium alloy which serves to prevent the build up of an electrostatic charge which will cause undesirable deflections of electron beam 40. The resis¬ tance of this coating is preferably high, and is prefer¬ ably 1,000 ohms when measured between the two ends of

ceramic cylinder 180, to minimize induced eddy currents. Stainless steel bellows 178 provides strain relieving mechanical connecting means to ceramic cylinder 180 to avoid the application of stress to the ceramic cylinder caused by, for example, mechanical misalignment.

As electron beam 40 is deflected in the desired scanning pattern across the face of target 50, the length of the electron beam path will vary. To compen¬ sate for this, the strength of the magnetic focus lens assembly 186 is preferably varied in synchronism with the scan to maintain the optimal size of focal spot 60. This is preferably accomplished by operating the static focus coil 185 at a fixed current. The small changes in strength of the field generated by dynamic focus coil 187 required to maintain the optimal size of focal spot 60 are achieved by modulating the current flowing in dynamic focus coil 187 in synchronism with the currents flowing in deflection yokes 188 and 190. The preferred means to control and drive the currents in the dynamic focus coil 187 and static focus coil 185 are discussed more fully in copending U.S. Patent Application Serial No. 08/386861, which has been incorporated herein by reference in its entirety.

X-rays are produced when electron beam 40 strikes target 50, which is preferably a circular plate with an active diameter of 25.4 cm (10 in) . A collimation grid 90 containing an array of x-ray transmissive apertures is preferably disposed between target 50 and multi-de¬ tector array 110. Target 50 and collimation grid 90 are discussed more fully in conjunction with the detailed description of Fig. 3.

Infra-red temperature sensor 192 monitors target 50 for excessive temperature conditions through viewing window 194 located in a wall of end bell assembly 266 opposite target 50. Excessive temperature conditions on target 50 may arise, e.g., if a malfunction causes elec¬ tron beam 40 to dwell for too long in one spot on target

50, instead of being scanned across its face. Infra-red sensor 192 preferably detects for excessively high temperatures by monitoring the amount of, or spectral shifts in, the luminosity of the face of target 50. The response time of sensor 192 is preferably of the order of one microsecond to avoid target burn-out.

Cooling jacket 196 and cooling plate 197 are preferably mounted on the exterior front wall and exterior perimeter walls of the end bell assembly 266, to remove heat generated by electrons which are back scattered from target 50 during normal operation of scanning beam x-ray source 10. Heat is removed from cooling jacket 196 and cooling plate 197 by use of a cooling fluid, preferably Fluorinert™, available from 3M Corporation, which is preferably circulated through an external heat exchanger (not shown) .

In the preferred embodiment, end bell assembly 266 is fabricated from stainless steel, conical in shape and double walled so that the cooling function can be achieved by circulating a cooling fluid in the space between the internal and external walls, thus the need for cooling plate 197 is consequently eliminated. The apex angle of the conical end bell assembly 266 pre¬ ferably conforms with that of the conical volume swept out by electron beam 40 while the radial dimensions of the inside wall of the cone are such as to provide preferably 1.2 cm spacing to the conical volume swept out by electron beam 40. This preferred shape reduces the internal surface area and the enclosed volume of end bell assembly 266 and the time required to evacuate the vacuum envelope assembly 176 to an acceptably low pres¬ sure.

Fig. 3 depicts a magnified diagrammatic view of the preferred target 50 and collimation grid 90 assembly. Target 50 preferably comprises a target layer 129 sup¬ ported by beryllium target support 130. A preferred construction of target layer 129 is a first layer of

niobium 51 approximately 1 micron thick applied to tar¬ get support 130 to which is then applied a second layer of tantalum 52 approximately 5 microns thick. The pre¬ ferred method of application for niobium 51 and tantalum 52 is by sputtering. Alternative methods include chemi¬ cal vapor deposition, evaporation and ion plating. Niobium layer 51 functions as a resilient layer which has a coefficient of thermal expansion between those of beryllium and tantalum to help prevent the formation of stress cracks in the tantalum layer 51, which may be caused by the high instantaneous temperature difference between the beryllium and the tantalum at focal spot 60 with consequent differential expansion between the tan¬ talum and the beryllium substrate which can cause crack- ing. In an alternative method for application of the target layer 129 to the target support 130, the coating process can be performed at an elevated temperature so that subsequent cooling produces a compressive stress in the target layer 129 to reduce the operating tensile stress in target layer 129 at focal spot 60 by an amount approximately equal to the initial compressive stress. Another embodiment is a layer of tantalum deposited directly on the target support 130. Yet another embodi¬ ment is a target layer 129 of an alloy of tungsten and rhenium. Still another embodiment is a target layer 139 of tungsten. In each of these embodiments an intermedi¬ ate layer of a resilient material such as niobium may be used. Tungsten, tantalum and tungsten-rhenium are pre¬ ferred materials for target layer 129 because they have high atomic numbers, making them efficient producers of x-rays, coupled with high thermal conductivity, high specific heat and high melting point. The thickness of target layer 129 is preferably selected to correspond with the distance traveled in the material by electrons of the highest operating energy. In an alternative embodiment, a lesser thickness is preferably used for target layer 129. In the first described embodiment

when the x-ray tube is operated at the low end of its operating range, for example 70kV, electrons which strike the target will not fully penetrate target layer 129, and the x-rays generated will then be attenuated as they pass through the remainder of the target layer 129. For a fixed electron beam power the x-ray flux at 70kV is about 30% of that at lOOkV so it is desirable to choose the thickness of target layer 129 based on the range of electrons in that material at 70kV in order to maximize the x-ray flux at 70 kV while accepting a slig¬ htly lower electron beam power to x-ray flux conversion efficiency at lOOkV. The conversion efficiency at lOOkV will nevertheless be greater than that at 70kV.

Beryllium is presently preferred for target support 130 because it possesses relatively high thermal conduc¬ tivity and it combines a low attenuation for x-rays with the high mechanical strength required to minimize the mechanical deflection of target support 130 caused by atmospheric and coolant pressures. The thickness of target support 130 is preferably about 0.5 cm.

Collimation grid 90 preferably comprises a circular array, 25.4 cm (10 in) in diameter of regularly spaced vertical columns and horizontal rows of apertures 140 with 166 apertures in both vertical and horizontal dia- meters. The total number of apertures 140 in collimation grid 90 is preferably about 21,642. The axis of each aperture 140 points towards the center of multi-element detector array 100 (Fig. 1) . While x-rays generated from focal spot 60 travel in all directions, collimation grid 90 provides a barrier which attenuates all those not directed towards detector array 110. The preferred collimation grid with alternative embodiments is described more fully in copending Patent Application Serial No. 08/386861. A cooling chamber 350 is preferably disposed between target 50 and collimation grid 90. Cooling chamber 350 is preferably 0.2 cm thick and may be adapted to carry

water, forced air or other types of cooling fluid. The presently preferred coolant is a liquid Fluorinert™ which is available from 3M Corporation. The coolant flows through cooling chamber 350 to absorb the heat dissipated by electron beam 40 as it strikes target 50. The coolant then passes through an external heat ex¬ changer where it is cooled before being recirculated to cooling chamber 350.

Fig. 4 is an exploded view of the components of high voltage terminal assembly 803. As described more fully in connection with Fig. 2, an approximate -70kV to - 120kV potential is preferably applied to high voltage terminal assembly 803 through spring 152, which is fit¬ ted into an indentation in the outer side of high volt- age terminal endplate 154. Circuit board 214 preferably contains the fiber optic communication circuits for the components of high voltage terminal assembly 803. Eight fiber optic communications cables are preferably con¬ nected to circuit board 214 through a plug-in feed- through assembly 166 in high voltage terminal endplate 154. The preferred fiber optic communications circuits are described more fully in copending Patent Application Serial No. 08/386861.

Because of the high voltage potential applied to the high voltage terminal assembly 803, an isolation trans¬ former 744 is preferably employed to supply power to the components within the high voltage terminal assembly 803. The secondary coil assembly 1271 of isolation transformer 744 is affixed to aluminum cylinder ring 226, which is shown bolted to the high voltage terminal endplate 154 and circuit board housing 212. A flat washer of conducting silicone rubber 288 is preferably compressed between one end of secondary coil assembly 1271 and the high voltage terminal endplate 154, provid- ing electrical conductivity between the two components. Similarly, another flat washer of conducting silicone rubber 290 is preferably compressed between the end of

secondary coil assembly 1271 and the circuit board hous¬ ing 212 to provide electrical conductivity between these two components. Circuit board housing 212 preferably comprises a deep drawn aluminum can with a large diame- ter axial hole formed in the can end plate. Contained within circuit board housing 212 is a circuit board stack comprising three circuit boards 206, 208 and 210. Circuit boards 206, 208 and 210 preferably contain all the electrical components which are necessary for opera- tion of electron gun 198. Specifically these circuit boards preferably contain a low-voltage heater power supply, a -2kV fixed power supply and a 0 to -2kV vari¬ able-voltage power supply. Each circuit board is circu¬ lar in shape and contains a solid "I" shaped aluminum heat sink 216 which protrudes from its surface. The circuit boards 206, 208 and 210 are bolted together to form a compact stacked assembly with electrical connec¬ tion between each board achieved by means of mating connectors mounted on each board. The fiber optic cir- cuitry located on circuit board 214 preferably connects to a connector on circuit board 210 via a ribbon cable which extends axially through the center of secondary coil assembly 1271 and through the hole in the end of circuit board housing 212. High voltage terminal sup- port member 202 is fabricated from aluminum in a conical shape with a rear flange fitted with two groups of three threaded rods 204 diametrically opposed on the rear of the flange. The three circuit boards 206, 208 and 210 are shown mounted to support member 202 by inserting threaded rods 204 through holes located on heat sinks 216. Circuit board housing 212 slides over circuit boards 206, 208 and 210 so that threaded rods 204 pro¬ trude through holes in the end of circuit board housing 212. Circuit boards 206, 208 and 210 are thus shielded from the effects of the intense electric field which exists on the external surface of circuit board housing 212. The assembly comprising circuit boards 206, 208,

210 together with circuit board housing 212 are prefera¬ bly held together by nuts applied to the ends of thread¬ ed rods 204. Referring to Fig. 2, high voltage terminal support member 202 is preferably bolted to flange 224 on electron gun 198.

Referring to Fig 4, electron gun 198 is preferably mounted within the frontal aperture of high voltage terminal support member 202. Referring to Fig. 2, a metal sleeve is preferably mounted within the central aperture of ceramic disc 128, and welded to this metal sleeve is a flanged vacuum tube 221. Flange 224 on electron gun assembly 198 is shown bolted to the flanged section of vacuum tube 221 with an intermediate copper gasket which provides a seal between the SF ₆ contained in electron beam source 112 and the high vacuum in vacuum envelope assembly 176. This arrangement allows for a simplified procedure for replacement of electron gun 198. Feedthrough leads 222 on electron gun assembly 198 pass through ceramic insulating disc 218 to make connec- tion to the internal electrodes of electron gun 198.

Electron gun 198 preferably comprises a heater coil embedded in electron-emitting cathode 220, with cathode 220 mounted behind control grid 200. The entire elec¬ tron gun structure is preferably supported from the feedthrough leads 222 on the vacuum envelope assembly 176 side of ceramic disc 218. Feedthrough leads 222 also provide electrical connecting means to the elec¬ trodes within the structure of electron gun 198. The presently preferred cathode 220 is a cylindrical piece of porous tungsten impregnated with low work-function materials which readily emit electrons. Such cathodes are known as dispenser cathodes and are available from Spectromat Inc. Employment of an impregnated tungsten cathode permits the use of a small diameter cathode since the electron beam current density obtainable from such a cathode is substantially higher than that from a pure metal emitter such as a tungsten filament. Because

the focal spot 60 is preferably small, the electron source is also preferably small. The embedded heater coil is energized by an electric current generated by a low-voltage heater power supply within high voltage terminal assembly 803 which flows through two of the feedthrough leads 222. The heater coil preferably rais¬ es the temperature of cathode 220 to approximately 1100- °C, which is the temperature at which the preferred cathode 220 emits the required electron beam current. These electrons are accelerated to an energy between

70keV and 120keV in the gap between electron gun 198 and anode 184 by the action of the negative high voltage applied to electron gun 198.

Control grid 200 preferably comprises a cylindrical electrode surrounding cathode 220 with an apertured end plate positioned slightly in front of the surface of cathode 220. The electron beam 40 emitted from cathode 220 can be varied in intensity by the application of a voltage to control grid 200, such voltage being of nega- tive polarity with respect to cathode 220. In the pre¬ ferred embodiment, application of -2kV from a fixed potential power supply in high voltage terminal assembly 803 through feedthrough leads 222 to control grid 200 completely inhibits the flow of electron beam 40. Ap- plication of a variable potential in the range 0 to -2kV to control grid 200 from a variable voltage power supply in high voltage terminal assembly 803 varies the inten¬ sity of electron beam 40 over the range of 0 to 60mA. X-ray source 10 is preferably operated in a pulsed mode such that electron beam 40 is pulsed on rapidly for a time period relating to the electron beam scanning mode. This is preferably achieved by means of two solid state switching circuits contained within the circuit boards 206, 208 and 210. Each switching circuit preferably comprises a series-connected string of field effect transistors which can be turned on and off by means of command signals conveyed through fiber optic cables 168.

Referring to Fig. 5, the components in circuit boards 206, 208 and 210 which generate heat, such as power transistors and voltage regulator components 217, are preferably attached to heat sinks 216. The three circuit boards are stacked and heat sinks 216 are clamped together by means of threaded rods 204. Heat dissipated in heat sinks 216 by components 217 is pref¬ erably conducted to high voltage terminal support member 202. Most of the heat will then be removed by convec- tion of the SF ₆ gas and thence to the outer walls of electron beam source 112. SF ₆ gas under pressure is the preferred heat exchange medium and natural convection forces are enhanced by circulation of the gas caused by the high electric field. Some of the heat from terminal support member 202 will also be removed by conduction through ceramic disk 128.

Fig. 6 is perspective diagram of a preferred isola¬ tion transformer 744, which supplies power for the com¬ ponents within high voltage terminal assembly 803. The secondary coil assembly 1271 of isolation transformer 744 is preferably located within high voltage terminal assembly 803. Because of the high voltage applied to high voltage terminal assembly 803, the primary coil assembly 1270 of preferred isolation transformer 744 is disposed coaxially around the secondary coil assembly 1271, physically separated from the secondary coil as¬ sembly 1271 by a distance of approximately 4cm(1.6").

Isolation transformer 744 preferably operates at a frequency of 60kHz although other operating frequencies can be employed. The potential at the secondary coil is preferably 30V R. .S. In the preferred embodiment, both the primary coil assembly 1270 and the secondary coil assembly 1271 each have a ferromagnetic core. The pre¬ ferred material for the ferromagnetic cores is ferrite, chosen for its low loss properties when operating at 60kHz, although other low loss materials such as com¬ pressed powdered iron can also be employed. Isolation

transformer 744 preferably operates with both primary and secondary coils resonant at the operating frequency by means of low loss capacitors connected across the coil connections. This improves the coupling between primary and secondary coils and eliminates the need for the 60kHz power source to provide the out-of-phase mag¬ netizing current. The resonating capacitors have a capacity of approximately O.lμF each.

The ferromagnetic core of secondary coil assembly 1271 is preferably formed of a cylinder of ferrite mate¬ rial. In the preferred embodiment, to reduce cost, the core is formed from a series of ferrite bars 230, pref¬ erably numbering twenty, with each bar abutting neigh¬ boring bars effectively forming a cylinder of ferrite within secondary coil 234.

As shown in Fig. 7, cylindrical coil form 232 preferably encircles ferrite bars 230, which are attached to coil form 232 by means of double sided adhesive foam tape 292. Coil form 232 is preferably formed of acrylic plastic although other electrically insulating materials with adequate thermal properties could be employed. Copper wire, preferably low loss RF wire known as Litz wire, is wound around the central part of the outer face of coil form 232 to form the secondary coil 234. In the preferred embodiment, there are 13 turns of wire in secondary coil 234 although the preferred number of turns is not germane to the essence of the invention and the actual number of turns depends on the particular usage requirements of the transformer. The number of turns depicted in Fig. 7 for secondary coil 234 is for purposes of illustration only and should not be considered the number of turns actually employed in the present invention.

Copper rings 236 and 238 are preferably placed around both edges of coil form 232. The coil form 232 is preferably longer than ferrite bars 230 by a small amount, preferably about 0.2cm (0.08"). Copper rings

236 and 238 form two short-circuited single-turn coils which completely encircle the upper and lower edges of coil form 232. In the preferred embodiment, copper rings 236 and 238 are formed from 0.125" OD copper tubing and have a diameter approximately equal to that of coil form 232. Copper rings 236 and 238 are attached to the upper and lower edges of coil form 232 by means of adhesive Kapton™ tape, available from Dupont. The entire assembly, comprising ferrite bars 230, coil form 232, copper rings 236 and 238, and the secondary coil 234 is then preferably wrapped toroidally with adhesive Kapton™ tape 300 to provide an electrically insulating protective barrier for secondary coil 234.

An electrostatic shield 240 is formed of insulated copper wire wound closely and toroidally around tape 300. The wire is preferably 24AWG copper magnet wire. The top and bottom surfaces of shield wires 240 are preferably treated to remove the insulation and expose the bare copper. A bead of solder 294 is preferably applied circumferentially around the bare copper wire surfaces to provide an electrical connection between adjacent wires at the top and the bottom. A conducting silicone rubber washer 288, preferably 0.15cm (0.060") thick, is preferably placed along the top surface of solder bead 294 and another identical washer of conducting silicone rubber 290 is preferably placed along the bottom solder bead.

Aluminum ring 226, containing bolt holes along the upper and lower edges, is preferably attached to the inner surface of secondary coil assembly 1271 by silicone rubber 228. Secondary coil assembly 1271 is attached to the rest of high voltage terminal assembly 803 by bolting aluminum ring 226 to circuit board housing 212 and the high voltage terminal endplate 154 (Fig. 4) . Conducting silicone rubber washers 288 and

290 are preferably compressed between the secondary coil assembly 1271 and the high-voltage terminal endplate on

one end, and the circuit board housing 212 on the other end, providing electrical conductivity between these components.

Cylindrical coil form 242 is the innermost layer of primary coil assembly 1270, as shown in Fig. 8. Coil form 242 is preferably formed of acrylic plastic although other electrically insulating materials with adequate thermal properties could be employed. Copper wire, preferably low loss RF wire known as Litz wire, is wound around the central part of the outer face of coil form 242 to form the primary coil 246. In the preferred embodiment there are 11 turns of wire in primary coil 246 although the number of turns is not germane to the essence of the invention and the actual number of turns depends on the particular usage requirements of the transformer. The number of turns depicted in Fig. 8 for primary coil 246 is for purposes of illustration only and should not be considered the number of turns actually employed in the present invention. Double sided adhesive foam tapes 296 and 298 are attached circumferentially to the upper and lower extremities of the outer face of coil form 242. Preferably attached to the outside of double sided adhesive foam tapes 296 and 298 is the primary coil ferromagnetic core material. In the preferred embodiment, the ferromagnetic core comprises a series of ferrite bars 252, preferably numbering 37, with each bar abutting neighboring bars effectively forming a cylinder of ferrite outside primary coil 246. Copper rings 248 and 250 are preferably placed around both edges of coil form 242. Coil form 242 is preferably longer than ferrite bars 252 by a small amount, preferably about 0.2cm (0.08"). Copper rings 248 and 250 form two short-circuited single turn coils which completely encircle the upper and lower edges of coil form 242. In the preferred embodiment, copper rings 248 and 250 are formed from 0.125" OD copper

tubing and have a diameter approximately equal to that of coil form 242. Copper rings 248 and 250 are attached to the upper and lower edges of coil form 242 by means of adhesive Kapton™ tape. The entire assembly, consisting of ferrite bars 252, coil form 242, copper rings 248 and 250, and the primary coil 246 is then preferably wrapped toroidally with adhesive Kapton™ tape 302 to provide an electrically insulating barrier for primary coil 246. An electrostatic shield 304 is preferably formed of insulated copper wire wound closely and toroidally around tape 302. The wire is preferably 24AWG copper magnet wire. The top and bottom surfaces of shield wires 304 are treated to remove the insulation and expose the bare copper. Beads of solder 306 and 308 are then applied circumferentially around the bare copper wire surfaces to provide an electrical connection between adjacent wires at the tope and the bottom. A conducting silicone rubber washer 310, preferably 0.15cm (O.060") thick, is placed along the top surface of solder bead 306 and a similar washer of conducting silicone rubber 312 is placed along the bottom solder bead 308.

The outer face of primary coil assembly 1270 is preferably affixed to central aluminum cylinder 118 with silicone rubber 254. Aluminum cylinder 118 contains bolt holes along its upper and lower edges and bolts to flanged aluminum cylinders 114 and 116 to form the outer wall of electron beam source 112. Conducting silicone rubber washers 310 and 312 are preferably compressed between the flanged portions of aluminum flanged cylinders 114 and 116 along both the top and bottom edges of primary coil assembly 1270. Conducting silicone rubber washers 310 and 312 thereby provide an electrically conductive path between flanged aluminum cylinders 114 and 116 and electrostatic shield 304.

Referring to Fig. 9, an alternative embodiment of primary coil assembly 1270 is shown. Like the preferred embodiment of Fig. 8, a cylindrical acrylic plastic coil form 348 preferably forms the inner structure of primary coil assembly 1270. Primary coil 350 is preferably wound around the outer face of coil form 348, preferably forming 11 turns. Two bands of adhesive double backed tape 352 and 354 affix a series of ferrite bars 356 to the upper and lower outer circumference of coil form 348. The outer face of primary coil assembly 1270 is preferably affixed to central aluminum cylinder 118 using silicone rubber 254. The number of turns depicted in Fig. 9 is for purposes of illustration only. The embodiment depicted in Fig. 9 employs an alternative electrostatic shield arrangement from that depicted in Fig. 8. This alternative electrostatic shield arrangement employs a resistive electrically conductive paint such as a colloidal graphite paint available from Acheson Colloids Inc., over the interior circumference 244 of coil form 348. As in the previous preferred embodiment, electrical connection to the graphite paint is made by conducting silicone rubber washers compressed against flanged aluminum cylinders 114 and 116. Fig. 10 illustrates the typical path of magnetic field lines which couple primary coil assembly 1270 with secondary coil assembly 1271 in the preferred isolation transformer 744. As depicted, currents induced in the copper eddy current shield rings confine the magnetic field and minimizes coupling into the various support structures, thereby minimizing excessive power losses in those support structures. Unlike conventional power transformers, magnetic field containment is achieved without a ferromagnetic return circuit. As shown in Fig. 10 the preferred isolation transformer 744 substantially confines the magnetic flux within the area defined by ferrite bars 230 and 252

and the copper rings 236, 238, 248 and 250. Ferrite bars 230 and 246, by virtue of their high permeability, provide a low reluctance path for the magnetic field so that the field travels preferentially in the ferrite bars rather than in the material on either side of the ferrite bars.

Eddy current shield rings 236, 238, 248 and 250 function as magnetic field clamps, substantially confining the magnetic flux Φ within the boundaries shown in Fig. 10. This field clamping occurs because the magnetic flux Φ induces eddy currents in the copper rings which in turn generate opposing magnetic fields which effectively "push in" magnetic flux field lines to within the above stated boundaries. Thus the typical path followed by magnetic flux Φ will extend from ferrite bar 252 towards ferrite bar 230 and will be curved with a bulge midway between copper rings 250 and 236 then returning from ferrite bar 230 towards ferrite bar 252 in a similar shaped path between copper rings 238 and 248. The ferrite bars together with the eddy current shield essentially function as a magnetic return circuit while maintaining physical separation between the primary coil assembly 1270 and the secondary coil assembly 1271. Two deflection yokes, fast yoke 188 and slow yoke

190 are preferentially employed to move electron beam 40 in the required scan pattern over the surface of target 50. Slow deflection yoke 190 preferably comprises saddle type X and Y deflection coils wound within the internal slots of a ferrite cylinder. Such a con¬ struction technique has been used for the deflection yokes used with television picture tubes. Fig. 11 shows a diagrammatic representation of a preferred fast deflection yoke 188. In Fig. 11, the x axis is defined as horizontal and the y axis is defined as vertical when Fig. 11 is viewed in its correct orientation. Y-step deflection coils 265 and 266 and X-step deflection coils

268 and 270 are toroidally wound with copper magnet wire in internal slots formed on the inside diameter of ferrite ring 286. The coils of fast deflection yoke 188 are preferably wound with fewer turns than the coils of slow deflection yoke 190 thus ensuring that the coils of fast deflection yoke 188 have substantially lower self inductances in comparison with those on slow deflection yoke 190. This lower self inductance of the coils on fast deflection yoke 188 makes it possible to effect small fast step changes in the amplitudes of the currents flowing in coils 265, 266, 268 and 270 with resultant rapid step changes in the position of electron beam 40 on target 50. The preferred circuitry to control and drive the current in the coils of the fast and slow deflection yokes are discussed more fully in copending U.S. Patent Application Serial No. 08/386861. The number of turns depicted in Fig. 11 is for purposes of illustration only and should not be considered the preferred number of turns. The deflection of electron beam 40 by deflection yokes 188 and 190 results in aberrations from ideal performance which increase in effect as the deflection angle of electron beam 40 increases. These aberrations cause focal spot 60 to depart from circularity as its distance from the center of target 50 increases. 45° stigmator coil 784 and 0° stigmator coil 786 are preferably employed to correct these aberrations. Currents supplied from an external source pass through stigmator coils 784 and 786 to modify the deflecting magnetic field configuration. The amplitudes and directions of these currents is programmed to maintain a circular shape for focal spot 60 as it scans over the face of target 50. 45° stigmator coil 784 and 0° stigmator coil 786 are preferably wound toroidally around ferrite ring 286 at the 0° and 45° positions.

The preferred circuits employed to control and drive the current in 45° stigmator coil 784 and 0° stigmator coil

786 are discussed more fully in copending U.S. Patent Application Serial No. 08/386861.

Referring to Figs. 12 and 13, means are provided to rotationally adjust the axes of the deflection yokes such that they are properly aligned in relation to the apertures of the collimation grid 90. Endplate 314, which is rigidly attached to the end bell assembly 266, contains two rotational support members 316 and 317 along its outer face, one on either side of the slow yoke 190. Rotational support member 316 contains a C- shaped section with an adjustment screw 318 inserted through the upper portion and adjustment screw 320 inserted through the lower portion of the C-shaped section. Slow yoke 190 is clamped between two identical alignment-clamps 326 and 328. Alignment-clamp 326 contains a flat rectangular tongue which extends outward between the upper and lower C-shaped portions of the rotational support member 316. Locking screw 322 extends through a groove 324 in the alignment-clamp tongue into a mating hole in the rotational support member 316. The adjustment screws 318 and 320 tighten to form contact with the upper and lower surfaces of the alignment-clamp tongue. A similar assembly exists on the other side of slow yoke 190 with respect to the other alignment-clamp 328. To effect the rotational adjustment of the slow yoke 190, locking screw 322 on alignment-clamp 326 and a similar locking screw on alignment-clamp 328 are loosened to allow free rotational movement of the Alignment-clamps 326 and 328. Adjustment screws 318 and 320, along with similar adjustment screws for alignment-clamp 328, are then adjusted to rotationally position the alignment-clamps 326 and 328, thereby effecting a corresponding rotational adjustment for the slow yoke 190 around the central ceramic cylinder 180.

A rotational support member 330 containing two rectangular protrusions extends and attaches through upper and lower rectangular grooves in alignment-clamps 326. Rotational support member 330 contains a C-shaped section with an adjustment screw 332 inserted through the upper portion and adjustment screw 334 inserted through the lower portion of the C-shaped section. A similar rotational support member 331 and locking screws 333 and 335 extend and attach to the other alignment- clamp 328.

Cylinder ring 338, which has the fast yoke 188 mounted along its interior surface, is formed with two rectangular adjustment plates 340 and 342 along its exterior surface. Rectangular adjustment plate 340 extends outward between the upper and lower C-shaped portions of the rotational support member 330. Locking screw 336 extends through a groove in the adjustment plate 340 into a mating hole in the rotational support member 330. The adjustment screws 332 and 334 tighten to form contact with the upper and lower surfaces of the adjustment plate 340. Adjustment plate 342 is similarly positioned between the upper and lower C-shaped portions of rotational support member 331. To effect the rotational adjustment of the fast yoke 190, locking screw 336 on adjustment plate 340 and a similar locking screw on the other adjustment plate 342 are loosened to allow free rotational movement of the cylinder ring 338. Adjustment screws 332 and 334, along with similar adjustment screws 333 and 335 for adjustment plate 342 are then adjusted to rotationally position the cylinder ring 338, thereby effecting a corresponding rotational adjustment of the attached fast yoke 188 around the central ceramic cylinder 180.

The magnetic focus lens assembly 186 can be positioned axially along the length of the vacuum envelope assembly 176 to regulate the minimum electron beam spot size on the target 50. Such positioning can

prevent damage to the target 50 from minimum electron beam spot sizes which are overly concentrated, which may burn the target 50. Positioning rod 274 extends from front endplate 138 to an endplate 314, which is rigidly attached to the end bell assembly 266. Five such positioning rods are preferably disposed equidistantly along the outside perimeter of the endplates 314 and 138. The magnetic focus lens assembly 186 is mounted between a front support plate 346 and a rear support plate 344. Preferably attached to the front support plate 346 are five rectangular clamps 276, each of which encircles a corresponding positioning rod 274. To position the focus coil structure 186, locking screws 278 on the clamps 276 are released allowing the focus coil structure to slide along the positioning rods 274. Once an optimal position is established, the locking screws 278 are tightened into a locking position. Magnetic focus lens assembly 186 can be moved radially to align the central magnetic axis of focus lens assembly 186 with the central axis of electron beam 40 when electron beam 40 is not deflected by yokes 188 and 190. Alignment of focus lens assembly 186 is effected by means of 4 set screws (not shown) , which protrude radially from threaded holes in plate 346. The inner ends of these set screws push against the outer diameter of the U-shaped magnetic circuit member. Turning these screws causes the magnetic circuit member to move in any radial direction with respect to plate 346. For purposes of illustration only, magnetic focus lens assembly 186 is shown as a solid in Fig. 12.

As discussed previously, the electron beam 40 is moved across the face of the target 50 in a predetermined scan pattern. Because of the collimation grid 90 employed in the preferred scanning beam x-ray imaging system, the electron beam 40 is preferably scanned in a "step" pattern. This step pattern is used to direct electron beam 40 to a spot on the target 50

that is on the axis of a specified collimator grid aperture 140 for a designated period of time, and then to rapidly move the electron beam 40 to another spot on the target 50 directly on the axis of the next specified collimator grid aperture 140. Electron beam 40 rapidly moves to the next target location to maximize the useful x-ray flux emitted through the collimator aperture.

The electron beam 40 is directed in this step pattern by the fast deflection yoke 188 working in combination with the slow deflection yoke 190. Within the slow deflection yoke 190, the X and Y deflection coils function in a conventional manner to apply a varying magnetic field such that the electron beam 40 is scanned in a sweep pattern across the target 50. The width and height of the sweep pattern is regulated by the current pattern applied to the X and Y deflection coils.

Within the fast deflection yoke 188, the X-step and Y-step deflection coils 264, 266, 268, 270 apply a rapidly moving magnetic field to modify the magnetic field generated by the slow deflection yoke 190. The combination of the magnetic fields generated by the fast and slow deflection yokes are such that the electron beam 40 is deflected in a step pattern across the target 50. Fast deflection yokes 188 are preferably employed because conventional slow deflection yokes designed to sweep the electron beam typically require a large voltage in order to change its current fast enough to generate the necessary step pattern, particularly in the preferred embodiment of the present invention where the electron beam is preferably stepped behind an 166 by 166 array of apertures with a scanning frame rate of 30 Hz. The coils in the preferred fast deflection yokes 188 are wound with shorter lengths and fewer turns than the slow deflection yokes 190, allowing fast current changes.

In a preferred embodiment, the electron beam 40 is deflected in a stepped raster scan pattern across the

face of the target 50, as depicted in Fig. 14. The preferred method to deflect the electron beam 40 in a raster scan pattern is diagrammed in Figs. 14A-F. Figs. 14A depicts a sample linear pattern applied to the X- deflection coils 280 and 282, producing a conventional X sweep of the target 50 by the electron beam 40. Fig. 14C depicts the sawtooth pattern applied to the X-step deflection coils 264 and 266, which produces the resultant step pattern as shown in Fig. 14E when magnetically combined with the X deflection pattern of Fig. 14A.

Fig. 14B depicts the pattern applied to the Y- deflection coils 276 and 278, to produce a conventional Y-sweep of the target 50 by the electron beam 40. As indicated in Fig. 14D, current is not applied to the Y- step deflection coils when scanning in the horizontal flyback mode since the period of time required for the electron beam 40 to "flyback" from the end of one horizontal row to the beginning of the next horizontal row gives the Y deflection coil sufficient reaction time to modify the current in its coil such that the electron beam is correctly deflected to the proper Y position.

In an alternate embodiment, the electron beam 40 is deflected in a stepped serpentine pattern across the target 50, as depicted in Fig. 15. The preferred method to deflect the electron beam 40 in a stepped serpentine pattern is diagrammed in Figs. 15A-F. Figs. 15A diagrams a sample pattern applied to the X-deflection coils 280 and 282, producing an X sweep of the target 50 by the electron beam 40. Fig. 15C depicts the sawtooth pattern applied to the X-step deflection coils, with a mirrored sawtooth pattern applied when the electron beam 40 begins scanning the next horizontal row, producing the resultant step pattern as shown in Fig. 15E when magnetically combined with the X deflection pattern of Fig. 15A. An alternate x-step pattern could comprise

the use of a negative sawtooth pattern during the return horizontal step period, as shown in Fig. 15G.

Fig. 15B depicts a sample current pattern applied to the Y-deflection coils 276 and 278, to produce a Y-sweep of the target 50 by the electron beam 40. The sawtooth Y-step pattern in Fig. 14D is applied when the scanning electron beam 40 reaches the end of a horizontal row, producing the resultant Y pattern shown in Fig. 14F when magnetically combined with the Y deflection coil pattern.

In another alternative embodiment, electron beam 40 is scanned in a stepped serpentine pattern as described in the previous embodiment but the Y-step coils are not used when the electron beam reaches the end of a horizontal row. The required y direction deflection of electron beam 40 is caused using the slow y coils in slow yoke 190. The greater time taken to achieve the step from row to row will typically result in a small reduction in efficiency of x-ray production. The size and shape of the current patterns depicted in Figs. 14A-F and 15A-F are shown for illustrative purposes only. The actual current patterns applied to the X and Y deflection coils and the X-step and Y-step deflection coils are dependant upon many factors, which may include the rate of movement of the electron beam, the amount of deflection already applied, the number of collimator apertures, the dwell time for each collimator aperture location, the number of turns for each coil, and the exact placement of the deflection coils. While embodiments, applications and advantages of the invention have been shown and described with sufficient clarity to enable one skilled in the art to make and use the invention, it would be equally apparent to those skilled in the art that many more embodiments, applications and advantages are possible without deviating from the inventive concepts disclosed and described herein. The invention therefore should only

be restricted in accordance with the spirit of the claims appended hereto and is not to be restricted by the preferred embodiments, specification or drawings.