US 3567940 A
Description (OCR text may contain errors)
Unite States Patent 72] Inventor Thomas W. Lambert Hales Corners, Wis.  Appl. No. 801,376  Filed Feb. 24, 1969  Patented Mar. 2, 1971  Assignee General Electric Company  CINERADIOGRAPHIC X-RAY TUBE GRID PULSING CIRCUIT EMPLOYING SERIES CONNECTED HIGH VOLTAGE SWITCHING TRANSISTORS 8 Claims, 2 Drawing Figs.
 U.S. C1 250/102, 250/65, 250/71  Int. Cl H05g 1/20, H05 g 1/60  Field of Search 250/102, 99; 315/209 (Inquired);/
 References Cited UNITED STATES PATENTS 3,428,809 2/1969 Daniels et al. 250/102 Primary Exa'miner.lames W. Lawrence Assistant Examiner-C. E. Church Attorneys- Ralph G. Hohenfeldt, Frank L Neuhauser and Oscar B. Waddell ABSTRACT: A light-emitting diode is pulsed at a preselected rate. Emitted light is conducted through a glass image conduit to a phototransistor which operates a solid state control switch at the same rate. The solid state switch controls conduction of a series connected string of transistors whose collectors swing from essentially zero voltage when the transistors are conducting to a high voltage when they are not conducting. The circuit features means for turning all of the transistors off in such manner that no transistor is subjected to overvoltage. The voltage pulses developed on the collector of the last transistor in the series may, for example, be applied to the grid of an xray tube to control its conduction intervals during high rate cinefluorography with an x-ray image converter. A signal that depends on image brightness is used to modulate the width of the pulses and, hence, the output of the x-ray tube so that each film frame has uniform exposure.
4 P/ P// P T g 22 a; F/O
PATENTEU "AR 2 |97l SHEU 2 OF 2 CINERADIOGRAPHIC X-RAY TUBE GRID PULSING CIRCUIT EMPLOYING SERIES CONNECTED HIGH VOLTAGE SWITCHING TRANSISTORS BACKGROUND OF THE INVENTION Cinetluorography involves converting a diagnostic x-ray image to a small bright visible image which appears on the output phosphor of an x-ray image converter tube. A cine camera views the phosphor and photographs the image at a high frame rate. The need for a high rate is exemplified in coronary arteriography where an x-ray opaque dye is injected in the blood vessels and it is desired to record the progress of its flow in the vessels and heart chambers. In such cases, the x-ray tube must be turned on when a film frame is stopped and turned off when the film is being transport d.
During cinerecording, the x-ray tube is turned on and off by pulsing its grid with a high voltage. Negative grid voltage as high as 5,000 volts may be required to cut off an x-ray tube which has a large focal spot and 150 kilovolts applied between its anode and cathode. In some cases, grid voltage is controlled by generating a bucking voltage on the secondary of a pulse transformer which is in series with a line from the bias voltage source. Since a pulse transformer cannot pass a DC component, it must be given time to recover before it can pass another pulse. Minimum recovery time turns out to be about 4 milliseconds in practice. The camera shutter is open and closed for equal time periods. At 120 frames per second the shutter is open for 4 milliseconds and closed for 4 milliseconds so it follows that 120 frames per second is the fastest frame rate attainable. The duty cycle, which is the ratio of pulse time to total frame cycle time is, therefore, limited to 50 percent. However, radiologists now prefer to record at much higher rates in order to stop motion properly.
Another disadvantage of using a pulse transformer is that grid voltage has slow rise and fall times. This has required the grid pulse to be initiated and the x-ray tube to be turned on prior to the time that the film frame is ready for exposure. Otherwise, the x-radiation intensity may not reach its peak soon enough and a frame will be underexposed. Premature turn-on and late turnoff of the x-ray tube results in the subject absorbing during each frame some radiation which does not contribute toward exposing the film. This wasted radiation loads the x-ray tube unduly and adds to the cumulative dose of the subject.
It has been proposed to use a series of silicon controlled rectifiers for switching the biasing source since a more nearly square biasing pulse can be attained in this way. However, several strings each including a large number of SCRs in series are required to switch high voltages. One string is needed to cause a voltage rise on a load resistor. Another string may be needed to supply the capacitive current to the x-ray tube cables. Two additional strings of SCRs must be provided for applying a blocking voltage to the other strings of SCRs so they turn off substantially at the same time. Such systems require many components and they are bulky and expensive to build. The minimum pulse width obtainable with SCR switching circuits is definitely limited as is well known.
It has also been proposed to switch the high grid biasing voltage with a series connected string of transistors which would permit sharp rise and fall times. Heretofore, this has not been feasible because of the difficulty of getting all series connected transistors to turn off and share the supply voltage equally. As is well known, the transistors that remain conductive have little voltage drop across them and those that turn off have a high voltage across them which causes avalanching and the system may go into oscillation.
SUMMARY OF THE INVENTION An object of the present invention is to overcome the above noted problems by providing a series transistor switching system that not only permits applying fast rise and fall time biasing pulses to the grid of an x-ray tube or other device, but
also facilitates modulating the width of the biasing pulses so that the total x-ray intensity delivered during each pulse can be precisely controlled. This means that image brightness and film exposure can be held constant although the image converter is scanned over regions of varying x-ray opacity during cinefluorography and although a dye of increasing opacity is entering the field of view.
A further object is to provide a high voltage switching system that permits any desired duty cycle and that permits frame rates and pulse rates that are higher than were heretofore obtainable.
A further object is to provide a fast high voltage transistor switch which has general application in the x-ray field such as for turning an x-ray tube on and off at a precise time in connection with phototimingand for turning an x-ray tube on and sustaining conduction for a long period of time and then turning it off such as in spot filming and fluorography.
Another object is to provide a high voltage transistor switch that has many uses such as for pulse modulating magnetron tubes in radar systems as well as in x-ray systems.
A basic object is to disclose a circuit that permits taking advantage of the fast switching characteristics of transistors for switching high voltages by series connecting transistors and providing means to assure that all transistors will turn on together and that they will turn off in quick succession and share the supply voltage equally.
Still another object is to attain better electrical isolation between a high voltage pulse circuit and a control circuit.
Briefly, these and other objects are achieved by employing several stages of series-connected high voltage transistors and a control transistor switching stage that controls the aforementioned stages. The control switch is triggered by the periodic flow of current through a phototransistor. The phototransistor is optically coupled to a light-emitting diode with a light-conducting glass conduit or rod which has high dielectric strength and provides good electrical isolation. A pulse generator that is both pulse width and pulse rate modu lated drives the diode at the chosen cine frame rate. A sensor continuously monitors average brightness of the output image of an x-ray image converter and, by means of an error signal in a closed loop system, the x-ray tube intensity is controlled at a preset level by continually modulating pulse width.
The transistor switching circuit features means for reverse biasing the base-emitter junctions of the transistors and thereby diverting the base storage current of the series connected transistors so they turn off in quick succession. Reverse bias increases the holdoff voltage of each transistor by a couple of hundred volts and contributes to short turnoff time.
A more detailed description of an illustrated embodiment of the invention will now be set forth in reference to the drawing.
DESCRIPTION OF THE DRAWING FIG. 1 is a block diagram of a diagnostic x-ray system incorporating the invention; and,
FIG. 2 is a schematic diagram of the new transistor system for switching high voltages.
DESCRIPTION OF A PREFERRED EMBODIMENT One use for the new grid pulsing system is in a cinefluorography system such as is depicted in FIG. 1. Here, an X-ray tube 1 projects its beam through a table top 2 and a patient 3 who is supported on the table tope. The differentially attenu ated x-rays enter an image tube 4 where they are converted to a bright optical image which appears on a fluorescent screen constituting an output phosphor 5. Electric power for operating the image tube is derived from a power supply 6.
Part of the light from the image on phosphor 5 may be deflected with a dichroic mirror 8 into a TV camera 9. The image may be visualized on a TV monitor 10. A spot film camera, not shown, may be substituted for the cinecamera to photograph the phosphor in a larger film size and in slower but precisely timed exposure intervals.
Part of the image light is also transmitted through dichroic mirror 8 to a cinecamera 7 which records at rates ranging from 1 to 240 or more frames per second in a practical embodiment. Faster cameras may be used, but persistence of the image tube phosphor is ordinarily the frame rate limiting factor in systems that use the new high speed switching system described herein. The operator sets the frame rate with a camera control 11 which also operates an electric signal source comprising a pulse generator 12 synchronously with and at the same rate as the camera 7. The width of the output pulses from generator 12 is controlled by a brightness regulator comprising an integrator and reference 14. Brightness control involves monitoring the average brightness of the image on output phosphor with a photomultiplier tube (PM) 13. The output current from PM13 is integrated to produce a voltage which is compared with a reference voltage that is set in accordance with the image brightness desired. The output error signal from integrator 14 which is used to control pulse width is not merely the difference between the integrated volt age value and the reference voltage. Rather, the pulse width controlling output error signal is the difference between the integrated and referenced voltages divided by the reference voltage signal expressed as a percentage of pulse width. This error signal is used to continually modulate the width of the pulses from generator 12 which results in control of the output image brightness.
Width modulated pulses at the selected cine recording frame rate drive a light-emitting diode (LED) 15. Light pulses from LEDlS are optically coupled to a phototransistor (PT) 17 through the agency ofa glass light conduit called rod 16 for brevity. Coupling a driving circuit to the driven circuit, which is about to be described, with the light rod 16 results in superior electrical isolation between the two circuits. Pulse transformers that couple the circuits must be very highly insulated which contributes toward their large size and high price.
Phototransistor 17 operates a solid state control switch 18 which is part of the new transistor grid voltage control switch 10. When a. light pulse occurs, solid state switch 19 effectively short-circuits the grid voltage supply so that the potential between grid 20 and cathode 21 of the x-ray tube is near zero in which case the x-ray tube conducts. in the absence of light pulses, solid state switch 19 becomes nonconductive and removes the short circuit so that the full negative bias voltage is applied to xray tube grid 20 in which case the x-ray tube stops conducting and no output image is formed on phosphor 5.
The x-ray tube power supply 23 is essentially conventional. For present purposes, a three-phase supply with low ripple is preferable. The high voltage that is applied between anode 22 and cathode 21 of the x-ray tube may be set with an x-ray con trol 24. The setting of this voltage depends on the penetrating power required. Control 24 also establishes the heating current and, hence, the electron emission of the filament or cathode 21. X-ray intensity depends on the temperature of the filament. Stepless control of filament current is not required in the present system. There may be one low current step for use during fluoroscopy when the tube is operated for long periods of time at low current. Two additional steps are provided for the cinefiuorographic mode when the filament is preferably brought to a higher temperature so that more intense radiation will be available. As explained earlier, the degree of film exposure during each frame is controlled by modulating the x-ray pulse width rather than by attempting to change the intensity of the beam by changing filament temperature during or between successive film frames.
PEG. 2 shows the details of the new transistor circuit for switching the grid bias voltage on x-ray tube 1. The circuit comprises a source 25 of DC bias voltage. The DC source 25 may furnish voltages ranging up to 5,000 volts when it is desired to control the grid of an x-ray tube which has a voltage of around X50 kilovolts applied between its anode and cathode and a large focal spot size. The positive terminal of DC source 25 connects to the top of a load resistor R1. The
bottom of R1 is connected to cathode 21 of the x-ray tube 1 through a resistor R17. The bottom-of R1 is also connected to several series connected transistor stages below in which the transistors of each stage are connected in the Darlington configuration in this example. The lowermost of the transistor stages connects to the negative side of DC source 25. it is evident, therefore, that when the transistor stages are nonconducting, the lowermost end of R1 is relatively positive in this example. When the transistor stages conduct, the lower end of R1 is at essentially the same potential as the negative side of the DC source 25. The grid 20 of x-ray tube 1 is connected directly to the negative side of the source. Hence, when the transistors are conducting and the lower end of R1 is essentially at the same potential as the negative side of the source 25, there is no substantial potential difference between grid 20 and cathode 21 and the x-ray tube 1 conducts. When the transistor stages are not conducting, a grid 20 is negative with respect to cathode 21 and the x-ray tube does not conduct.
As stated earlier, the high voltage transistor switching circuit shown in FIG. 2 is turned on and off with a phototransistor 17 that receives light pulses at a controlled rate and width from a light emitting diode 15. Phototransistor l7 conducts when it is exposed to light. As shown, phototransistor 17 controls a fast electronic switch 26 which is part of the control switch 18 that is encompassed in the dashed line rectangle. This switch 26 may take many known forms so it is shown in block form. Switch 26 is connected to the base of a transistor Q1. Q1 in this circuit is conducting when phototransistor 17 is not exposed to light. When the phototransistor receives a light pulse, switch 26 effectively connects the base of transistor Q1 to its emitter or the negative side of the source 25 through a connection 27. This removes the forward bias from the base and emitter of transistor Q1 and causes its collector to go positive. The collector resistor R28 of Q1 connects through switch 26 to the positive side of the capacitor C29 which has large capacity and an essentially constant voltage on it by virtue of it being in parallel with a zener diode 30. Capacitor C29 serves as a regulated source of voltage for the collector of Qll. When there is no light pulse, O1 is conducting, its collector is at essentially the potential of its emitter and the base-to-emitter circuit of a pair of Darlington connected transistors Q2 and O3 is not forward biased, so transistors Q2 and Q3 are nonconducting. When a light pulse occurs, Q1 stops conducting, its collector goes positive and both transistors Q2 and Q3 become forwardbiased. Forward-bias current is derived from capacitor C29 and is conducted through resistor R28, the base-emitter junctions of Q2 and Q3 and to the negative side of the source through forward-biased diodes D1 and D2. When Q2 and Q3 are forward biased, they begin conducting in the usual manner from their collectors to their emitters. This drops the potential on the emitters of Q4 and O5 in the next stage so that they become forward-biased by reason of the positive voltage at point 32 of a resistive voltage divider R2R4 driving current through base resistor R5 and the bases of Q4 and Q5 and to the negative line through Q2 and Q3 which are then conducting. All other transistor stages turn on in the same manner within a couple of microseconds. The bottom of load resistor R1 and cathode 21 of the x-ray tube are then at the same potential as grid 20 and the x-ray tube emits x-rays for the duration of the light pulse. From the descriptions thus far, it should be evident that electronic switch 26 and transistors Q1, Q2 and Q3 together with their associated circuitry act as a control switch stage 18 for turning the other transistors on and off in response to light pulses or other signals.
The number of additional Darlington switching stages such as Q4, Q5, Q6 and O7 is dictated by the x-ray tube grid bias voltage that is to be switched and the holdoff voltage of the individual transistors. In a practical embodiment, there are eleven such Darlington stages for switching up to 5,000 volts. Each transistor has a rated holdoff voltage of 450 volts. The Darlington configuration is used to obtain the desired current carrying capacity. I
The resistor voltage divider network comprising series connected resistors R2, R3 and R4 has as many resistive steps as there are transistor stages such as the one comprising Q4 and Q5. The use of additional resistive steps and stages is suggested by the dashed lines 31 between the second and third stages counting from the bottom in FIG. 2. The voltage of DC source 25 divides equally across these resistors. The voltage at a point such as 32 on the top of R2 supplies forward-biasing potential for the base-to-emitter circuits of transistors Q4 and Q5 mentioned above. Bias current to these transistors is limited by an appropriate resistor such as R5. Since the total voltage at a point such as 33 in the upper stage is higher than the voltage at point such as 32, it is necessary to provide base resistors of higher value for the upper stages. Thus, R6 and R7, the base resistors of Q6 and Q7, have a higher value than R5. The object in any case is to bias the base circuits of all transistors with equal current. Since the various transistor stages are connected in series and across the DC source 25, the voltage also divides equally from the collectors in one stage to the collectors of the transistors in adjacent stages with each stage acting as an emitter-follower. Equal potential gradient across the transistors in the various stages is assured by another voltage divider comprising resistors R8, R9, R10 and R11 which compensate for collector leakage current. When the first stage comprising transistors Q2 and O3 is turned on, all the other stages go on and the voltage between the respective collectors and emitters approaches zero. When the first stage is turned off, all other stages go off in quick succession according to the inventionand the divided voltage value appears between the respective collectors and emitters.
When a light pulse occurs, transistors Q2 and Q3 in the first stage or control switch 18 turn on as has been explained above. When this happens, the emitters of transistors Q4 and O5 in the second stage are effectively connected to the negative side of the line as explained. This results in their base circuits being forward-biased from potential derived at point 32. This forward-bias current through transistors Q4 and Q5 renders their collector-to-emitter paths conductive. These paths have low impedance even though they include the impedances of Q2 and Q3. When the second stage turns on it, of course, connects the emitters of the third stage transistors Q6 and Q7 to the negative side of the DC source 25 and permits their bases to be forward-biased. This turnon process is repeated in quick succession for as many stages as there are in the switching circuit.
There is no great difficulty in getting series connected transistors to turn on essentially simultaneously. When one stage turns on, it merely imposes a higher potential across the other stages and the other stages become conducting. Thus, the collector-to-emitter path of the transistors will not breakdown due to overvoltage. However, transistors are seldom connected in series to increase voltage capability because of the difficulty involved in forcing the voltage across each transistor to be equal when the transistors are being turned off. When the first stage tends to turn off, its collector voltage tends to rise toward supply voltage which is far in excess of what the transistor is capable of sustaining. As is known, even though the base-to-emitter bias is removed from a transistor, it remains on until the base storage chargeis removed.
in the new circuit, a path is provided for diverting the base storage current from each transistor when the transistor'is being turned off. This diverting path includes a diode such as D3 which is connected between the emitter and base of 05 so that when the diode conducts diverted current, the baseernitter junction is reversebiased and turn off of the transistor is assisted. it is known that a transistor has a capability for withstanding the highestcollector voltage when its base-toemitter circuit is reverse-biased. in the circuit under discussion, collector current from any stage that is still on is conducted through diodes to a capacitor. This causes reverse bias and prevents collector current from flowing through lower stages during turnoff.
Assume that in the absence of a light pulse Q2 and Q3 tend to turn off. This will cause the voltage on the emitters of transistors Q4 and Q5 and on the collectors of Q2 and Q3 to rise. The rising voltage will tend to breakdown transistors 02 and Q3. According to the invention, however, the voltage rise on the collector of Q3 and emitter of Q5 that tends to develop during turn off, causes a pair of diodes D3 and D4 to be forward-biased in which case, current flows from the emitters of Q4 and Q5 serially through diodes D3 and D4 and another diode D5 to a capacitor C1. Capacitor C1 is connected in series as a dividerwith additional capacitors such as C2 and C3 and the series group is connected along with capacitor C29 across the DC source. The normal voltage across capacitors C1 to C3 is the same as the divided voltage across the collectors and emitters of the transistors in the individual stages. These capacitors are isolated from the base circuit of the transistors by a reverse-biased diode such as D5. Thus, when diodes D3 and D4 are forward-biased, so is diode D5, in which case current flows from the emitters of Q4 and O5 to capacitor C1. The forward-bias current through diodes D3 and D4 results in a voltage drop of about six-tenths of a volt across each of these diodes. The direction of the voltage drop is effective to reverse-bias the base-to-emitter junctions of Q4 and O5 in which case the transistors not only turn off abruptly, but are put in a reverse-biased state and they are able to withstand the highest sustaining voltage.
The same turn off process occurs consecutively in the other transistor stages. For instance, when the voltage on the collectors of Q4 and Q5 tends to rise, so does the voltage on the emitters of Q6 and Q7 and .diodes D6 and D7 become forward-biased. This amounts to the bases of Q6 and Q7 being reverse-biased by reason of current flowing through D6, D7 and DS-to capacitor C2. Hence, it is seen that a voltage rise on the collectors of any stage will be prevented by forward biasing the diodes which shunt the transistors in any adjacent stage.
Capacitors C1, C2, and C3 each have a charge equalizing resistor such as R12, R13 and R14 connected respectively in parallel with them. The RC combinations are chosen so that the average potential on the capacitors remains essentially the same when they receive a rapid succession of diverted current pulses. The resistor is of such size that its average current settles on the average current coming into the capacitor. in other words, the average collector current times the resistance value equals the increase of voltage on the capacitor. in a practical embodiment, a 7 volt rise on the capacitors occurred when there was a long series of switching cycles.
Transistors Q2 and O3 in the first or control switching stage 18 are also reverse-biased during turn off. This may be seen by considering that when the light pulse disappears, Q1 switches to a conductive state instantly. At this time, the emitters of Q2 and Q3 are just slightly more positive than the negative line because of the total of about 1.2 volts drop across series connected diodes D1 and D2 resulting from the transistor-collector current flowing. in parallel with diodes D1 and D2, are a pair of series connected diodes D9 and D10. The two sets of series connected diodes have their cathodes connected together through transistor Q1, when it becomes conducting, so collector current flows from the emitters of Q2 and Q3 through the respective diodes D10 and D9 and D1 and D2 to the negative side of the source through Q1. Current flow through diodes D9 and D10 imposes a reverse-bias voltage on Q2 and Q3 and cuts these transistors off abruptly.
it is known that transistors such as Q2 have a very small leakage current between their collectors and bases to the emitters when they are subjected to high voltage. Because of the sensitivity of the instant high voltage transistor circuit, it is desirable not to have any leakage current flowing through the base-emitter junction or the transistors may turn on accidentally. To preclude this, the base and emitter junctions of the transistors are shunted by resistors such as R15 and R16. This causes leakage current to take the path from the collector to the base and through the shunting resistors without passing through the emitters. The transistors in the other stages have similar shunting resistors such as R18, R19, and R21, R22.
in this embodiment, there is a surge current diode D11 connected between grid of x-ray tube 1 and its cathode 21. The purpose of this is to provide a direct circulating path in the event there is a high voltage breakdown between the anode and grid of the x-ray tube 1 so that fault current will not enter the switching circuit. Diode D11 must be capable of withstanding the full grid biasing voltage. Another diode D12 is connected in parallel with load resistor R1 to protect it against the heavy reverse currents that wouldflow if there were a breakdown in the x-ray tube. Resistor R17 which is connected between the load resistor R1 and the x-ray tube cathode 21 is for limiting the surge current that flows when the series transistors turn on. The surge current results from capacitance of the x-ray tube cables which charge when the bias voltage is applied to the x-ray tube and discharge when the series connected transistors become conductive.
In FIG. 2 there is shown a conductor 34 between the midpoint of the DC source voltage and the midpoint of the resistor voltage divider network R2-R4. This conductor provides a path for some of the biasing current that flows to the bases of the various transistors through resistors such as R5. Other wise, other resistors in the network such as R4 would have to carry the sum of the biasing currents ,for all of the transistors and resistor R4 would have to have a high wattage rating.
Commercial phototransistors 17 are provided with an external base terminal so that the base-to-emitter junction can be forward-biased if desired to thereby increase the sensitivity of the phototransistor. External biasing is omitted in this case because it was found that the phototransistor becomes extremely sensitive to noise if there is external biasing.
The new transistorized switching circuit of FIG. 2 has demonstrated a capability of permitting cinerecording at rates as high as 500 frames per second, if desired, which is well above the highest frame rate of presently available cameras. The pulse rate may be in the kilocycle range if the application requires. The new circuit is also distinguished switching so fast that it permits effective radiation pulse widths ranging down to zero if desired. The triggering light pulses and the corresponding x-ray pulses are relatively square. No excess radiation is administered to the patient by turning on the x-ray tube prematurely as was required heretofore in order to assure that its output will peak before a frame is exposed. Constant film density is obtained by detecting the average brightness of the output phosphor in the image converter tube and using the signal so developed to modulate the width of the x-ray pulses. This results in better exposure control than is obtainable by attempting to control x-ray tube filament'current or applied voltage.
Although construction and use of the new transistor switching circuit is illustrated in connection with a grid controlled x-ray tube, those versed in the electrical arts will appreciate that the new circuit may be used in many other high voltage switching applications which have not heretofore been possible because of the difficulty of compelling series connected transistors to share source voltage equally during turn off.
l. A transistor switching system for controlling the bias voltage between the grid and cathode of an x-ray tube comprising:
a. terminals that are adapted to be connected to a DC source;
b. a resistor voltage divider connected across said source terminals;
. c. a load resistor means having one-end connected to the positive source terminal; I
d. a plurality of series connected transistor stages each of which has a base, an emitter and a collector, the collector of one being connected to the other end of the load resistor and the emitter of the one being connected to the series connected collector-emitter paths of the others;
e. a control switch connected between the last emitter in the series and the negative source terminal;
f. base resistors connected between successive intermediate points on the resistor divider and the respective transistor bases whereby to supply forward-bias current through the base-emitter junctions of the transistors when said control switch is conducting to turn the series connected transistors on and thereby make the potential of the other end of the load resistor substantially equal to that of the negative source terminal;
g. a voltage divider comprising series connected capacitors that are connected across said source terminals;
h. a first diode connected between each transistor base and a corresponding intermediate point on the capacitor voltage divider, said diodes being biased oppositely at the base-emitter junctions;
. a second diode connected in parallel with each baseemitter junction and biased in the direction of the first diode and being in series with it to provide a conductive path between each emitter and its associated capacitor;
j. the turning off of said control switch causing successive increased potentials on the transistor-emitters, beginning with the stage next to the control switch, which causes collector-emitter current from the stage to divert through the second and first diodes into the said capacitor whereby to reverse bias the base-emitter junctions of the series connected transistors in succession during turn off;
k. a discharge resistor in parallel with each capacitor;
. a collector leakage current resistor shunting the emitter and collector of path of each transistor, said leakage current resistors being series connected to each other and connected between said source terminals;
m. turnoff of said series transistor stages causing the voltage to rise positively on said other end of the load resistor with respect to said second source terminal;
n. an x-ray tube having an anode, a cathode and a control grid; and
o. the cathode of said x-ray tube being connected to the other end of said load resistor means and said grid being connected to the cathode when said transistor switches are not conducting to thereby turn off the x-ray tube.
2. The invention set forth in claim 1 including:
a. a phototransistor controlling conduction of said control switch;
b. a light source that is operated synchronously with the desired operating intervals of the x-ray tube, said light source being optically coupled with said phototransistor; and
c. an electric signal source adapted to energize said light source for timed intervals to thereby turn said control switch on and off. v
3. The invention set forth in claim 2 wherein:
said light source is a light-emitting diode and a light eonduit couples the same to said phototransistor.
. The invention set forth in claim 2 including:
. an x-ray image converter having an output phosphor on which there is an optical image when the x-ray tube is energized;
b. a camera optically coupled with said phosphor;
c. a camera control adapted to operate the camera at a selected recording rate; and
d. the said electric signal source being adapted to produce an output signal synchronously with operation of the camera to thereby turn said control switch on and off as aforesaid and to thereby control both conduction of said transistor means and the bias on said x-ray tube grid.
5. The invention set forth in claim 4 including:
a. detector means adapted to produce a signal proportional to the brightness of the image on the phosphor;
b. means that respond to the signal from said detector by producing a second signal that represents a percentage deviation from reference brightness; and
c. said electric signal source comprising a pulse generator that is pulse rate modulated in response to the frame rate of the camera and is continually pulse width modulated in response to said second signal, whereby the exposure of the consecutive film frames is held substantially uniform.
6. The invention set forth in claim 1 including:
a. an x-ray fluorescent screen;
b. a camera optically coupled with the screen;
c. a photodetector optically coupled with the screen and producing a first electric signal depending on brightness of the screen;
d. means integrating said first electric signal with respect to time and producing an output signal that is related to brightness;
e. an electric signal source whose output signal is controlled by said integrated signal;
f. a light-emitting diode energized by the output signals from said electric signal source; and
g. a phototransistor optically coupled with said lightemitting diode said phototransistor controlling said control switch.
7. The invention set forth in claim 1 including:
a. a phototransistor controlling said control switch;
b. an electrically energized light-emitting device optically coupled with the phototransistor; and
c. an electric signal source for selectively energizing and deenergizing said light-emitting device to define x-ray exposure intervals.
8. The invention set forth in claim 7 including:
a. a cinerecording camera that is operable at a preselected frame rate for recording an optical version of an x-ray image;
b. the said electric signal source being a pulse generator that is adapted to have its pulse rate modulated in accordance with the frame rate of the camera and the width of its output pulses modulated in accordance with the brightness of said optical image; and
c the x-ray tube being turned on and off in synchronism with the camera.