PATENTSCOPE will be unavailable a few hours for maintenance reason on Tuesday 19.11.2019 at 4:00 PM CET
Search International and National Patent Collections
Some content of this application is unavailable at the moment.
If this situation persists, please contact us atFeedback&Contact
1. (WO1985004545) BALLAST AND CONTROL UNIT FOR ELECTRIC DISCHARGE LAMP
Note: Text based on automatic Optical Character Recognition processes. Please use the PDF version for legal matters

BALLAST AND CONTROL UNIT FOR ELECTRIC DISCHARGE LAMP
Technical Field
This invention relates generally to electric discharge lamps and particularly to ballast units and control units as used therewith.
Background Art
Electric discharge lamps, such as fluorescent lamps, operate by applying an electric current through a gas such that at least some of the gas atoms become ionized. When enough atoms are ionized, the gas becomes an electric conductor and radiation results.
Such electric discharge lamps typically require ballast units to provide the necessary starting voltage and to keep the current within proper limits. Such lamps further generally require a temporary starting circuit to provide current to heat the electrodes when the lamp has first been turned on. Such ballast and control units are subject to a number of problems, including radio frequency interference, noise, flicker and the like, and further do little to efficiently control the operating temperature or energy utilization of the lamp. Neglect of these factors can affect the anticipated life of both the lamp and the ballast unit.
Further, certain high efficiency electric discharge lamps require an even greater initial application of voltage than would ordinarily be required with ordinary electric discharge lamps in order to initiate the desired radiation process. This initial expenditure of power, however, becomes more than offset during the highly efficient operation of the lamp (savings of ten percent and more can he expected). Although these high efficiency lamps are desirable from this standpoint, many prior art ballast and control units are not capable of providing the proper initial starting voltage to ensure that the lamp will be started.
Also, to ensure efficient and proper operation, the ballast unit should provide a degree of power factor control to ensure that lamp current flows during the entire applied AC sine wave cycle. This provides maximum light output efficiency for a given power input to the lamp. The techniques for power factor correction as used in the past for fluorescent lamps have generally been complex in construction and operation. Moreover, such techniques have resulted in circuits that are relatively expensive to produce and maintain.
In addition, the electric discharge lamps in any unit must be replaced from time-to-time by maintenance personnel. There exists a risk of electric shock to such persons when performing this task. Prior art devices do little to aid in protecting such personnel from this danger.
There therefore exists a need for a ballast and control unit for electric discharge lamps that combines high efficiency with good starting
characteristics such that even high efficiency lamps can be utilized therewith. Further, such a unit should include protective mechanisms to avoid exposing maintenance personnel to unnecessary dangers when changing lamps, and should provide efficient power factor control.
Disclosure of Invention
These needs and others are substantially met through provision of the ballast and control unit for electric discharge lamps disclosed in this
specification. This ballast and control unit includes generally a phase differential control unit, a lamp starting unit, a heater control unit, and a
disconnected lamp sensor and current limiter unit.
The phase differential control unit operates to ensure that input voltage and input current are substantially in phase such that a power factor of 0.9 or greater can be realized. This contributes to an effective use of available power and an efficient and economical powering of the lamps. In addition, the phase differential control unit also aids in
maintaining a desired relationship between output voltage and output current as supplied to the lamps in order to maximize light output from the lamps with respect to the power being applied thereto.
These benefits are obtained in part through use of a ballast having a plurality of ferrous layers laminated together to form an H-shaped core having wound bobbins fit over either end thereof. So
constructed, the H-shaped core forms two transformer couplers, with the primary of the first transformer being connected to the power input, the secondary of the first transformer being connected to the primary of the second transformer, and the secondary of the second transformer being operably connected to the lamps.
The lamp starting unit includes circuitry for sensing when the lamps are on, and additional circuitry to alternately apply full starting voltage to first a single lamp, and then to two or more series connected lamps until the lamps are on. Through provision of this unit, a sufficiently high voltage can be applied to the lamps to ensure starting of even high efficiency lamps.
The heater control unit contains circuitry to ensure the supply of heater voltage to the lamps until the lamps have been turned on. Once the lamps have been turned on, the sensing circuitry of the lamp starting unit causes the heater control unit to turn off, thereby saving power and contributing to more efficient operation of the combined system.
Finally, the disconnected lamp sensor and current limiter unit includes circuitry to detect when one or more lamps are electrically disconnected, and
additional circuitry to limit the applied voltage (and hence the available current) that may be applied against a careless maintenance person. Through use of this unit, full voltage and current will be supplied to the lamps unless and until one or more lamps become disconnected. Once this disconnection occurs, the supply voltage and current will be limited to well within safe limits to guard against inadvertent injury to a person changing the lamps or otherwise
investigating the status of the system.
Brief Description of the Drawings
These and other attributes of the invention will become more clear upon a thorough review and study of the following description of the best mode for carrying out the invention, particularly when reviewed in conjunction with the drawings, wherein:
Fig. 1 comprises a block diagram depiction of the invention;

Fig. 2 comprises a perspective exploded view of a pair of inductors on an H-shaped core;
Fig. 3 comprises a sectioned side elevational view of the core;
Fig. 4 comprises a schematic view of a first control circuit; and
Fig. 5 comprises a schematic view of a second control circuit in accordance with the block diagram of Fig. 1.
Best Mode for Carrying Out the Invention
Referring now to the drawings, and in particular to Fig. 1, the invention may be seen as depicted generally in block diagram form by the numeral 10. The invention (10) serves to operably connect a power source (11 ) (such as a 277 volt AC (RMS) power source) to one or more electric discharge lamps (12). which lamps (12.) are typically connected in series with one another. In one embodiment, the invention (10 ) includes generally an input unit (13), an output unit (14), phase differential control unit (16), a heater control unit (17), a lamp starting unit (18), and a disconnected lamp sensor and current limiter unit (19). Each of these units will be described in more detail in seriatim fashion below.
Referring first to Fig. 2, this invention makes use of an H-shaped core as denoted generally by the numeral20. This core (20) allows maximum winding area for a first inductor (21) and also serves to reduce series resistance in the lamp current path.
This core (20) comprises a stack of lamination that are arranged so that the core has a height of about .625 inch. A typical stack may have pieces that are aeach .014 inch thick. The core (20) has a pair of relatively long legs (22) and (23), a pair of relatively short legs (14) and (16), and a cross piece that provides for mutual coupling between the
inductors formed thereon.
The rst inductor (21) comprises a pair of windings (27) and (28) on bobbin bodies and in series with each other, the windings (27) and (28) being mounted on the long legs (22) and (23), respectively. To complete the flux paths in these legs (22) and (23 ), a smaller H-shaped core member (29) couples to the ends of the legs (22) and (23) such that the bridging member (31) of this smaller core (29) bridges the gap between the inner faces (32) and (33) of the longer legs (22) and (23).
A second inductor (34) also includes a pair of windings (36) and (37) on bobbin bodies and in series with each other that mount on the shorter legs (2A) and (2.6), respectively, of the main core (20). (The common point (38) between the two inductors (21) and (34 ) can be seen as an electrical connection in Fig. 3). To complete the magnetic flux path between the shorter legs (24) and (26), a second smaller H-shaped core member (39) is provided between the inner end faces (41 ) and (42) of these legs (24) and (26) (Fig. 2).
Lamination fasteners (43) and (44) are used to hold the smaller core members (29) and (39) in place. The first fastener (43) may be comprised of non-magnetic material and the second fastener (44) can be comprised of magnetic material.
Fig. 3 shows the two inductors (21) and (34) mounted in operative positions on the H-shaped core (20 ). The first of the smaller ocre members (29) has a magnetic gap (46) formed between itself and the main core (2θ) to prevent core saturation. Typically, this gap may be about .030 inch. The smaller core member (39 ) for the second inductor (34( has no such gap and forms an interference fit to allow maximum flux coup]ling. THe coils of both inductors (21 ) and (34) are wired to allow series aiding of magnetic fields. The core (20) can be formed of grain oriented silicon steel to reduce core losses. Because of the much longer flux path associated with the first inductor (21), the magnetic flux will be distributed over a much larger core length, resulting in less localized saturation and thereby less core loss or heating.
The second inductor (34) is used in conjunction with the first inductor (21) to adjust the power line power factor independently of the lamp power factor. THis advantage allows for maximum system efficiency. The unique H core configuration results in improvement of system efficiency of up to 15% over conventional core/coil type ballasts.
Referring to Fig. 3, the input unit (13) includes input terminals (47) for connection to the power source and a thermal switch (48) such as a Texas Instruments part no. 7AM023A5-134 to act as a
mechanical thermal cutout switch. With this switch (48 ) the power supply line will be automatically opened at 80 degrees centigrade (176 degrees Fahrenheit) in the event of ballast failure. The input unit (13) also includes a 91 ohm fuse action resistor (50) connected in line between the power source terminals (47) and the heater transformer described below. This resistor (50) serves to protect the ballast in the event that the heater transformer develops an internal short and further serves to prevent excessive temperature rises in the heater transformer during various failure modes.

The various connections to the primary and secondary windings of the first and second wound bobbin assemblies (21 and 34) are made through
appropriate output and input ports (49) in conformance with well known prior art techniques.
The H core dual coupler inductor (20) has the two primary windings of the first wound bobbin assembly (34 ) connected to the power source terminals (47). In addition, a winding tap (51) connects through a 1N4001 diode (52) to a 220 microfarad 16 volt electrolytic capacitor (53). The diode and capacitor (52 and 53) cooperate to provide a 10 volt power supply (54) for use in the circuit as appropriate.
The secondary windings of the second wound bobbin assembly (21) are connected together and to the lamps (|2) through a pair of parallel connected 1.35
microfarad 600 VDC capacitors (56 and 57 ). Finally, a central tap between the two wound bobbin assemblies (21 and 34) connects to the disconnected lamp sensor and current limiter unit (19) as described in more detail below.
With continued reference to Fig. 5, the heater control unit (17) will now be described.
A .22 ohm 1 watt resistor (58) connects between the power source terminals (47 ) and one side of the lamps (12). So positioned, this resistor (58) can sense lamp arc current, and apply a voltage to the input of an LM324 operational amplifier (59) through a 10K ohm resistor as provided through use of a DIP package (61). The remaining input to this operational amplifier (59) connects to the opposing side of the current sensing resistor (58) through a 10K ohm resistor (6dd) as contained in the DIP package (61) described above.

A 249K ohm feedback resistor (62) connects between the output of the operational amplifier (59) and the inverting input thereof. The output also connects through a series connected pair of 10K ohm resistors (61b and 61c) to the inverting input of a second LM324 operational amplifier (63) and to a 22 microfarad 6 volt electrolytic capacitor (64) that serves to integrate the input to the second
operational amplifier (63).
The remaining input to the second operational amplifier (63) connects through a 10K ohm resistor (66 ) to the positive 10 volt source (54) and through a 1N4148 diode (67) to the lamp unit (12). Finally, the output of the second operational amplifier (63) connects through a 330 ohm resistor (68) to the gate of a triac (69) (as provided through use of a
L6004F31) that connects between the low side of the power source terminals (41) and one side of the primary winding of the heater transformer (70) as described below.
The heater transformer (70) has one side of its primary winding connected to the power source
terminals (47) and the remaining side connected to the triac (69) as described above. This transformer (70) has three secondary windings. One (70a.) connects across the heater electrodes of the first lamp (71), one (70b) connects across the heater electrodes of the second lamp (72), and the last (70c) connects to a capacitor and diode network (13 and 74 ) (as provided through use of a 470 microfarad capacitor and a 1N4001 diode) that serves to provide a positive 5 volt source for use as necessary in this invention.
So positioned, the triac (69) can control the flow of current through the heater transformer, and hence control whether the heater transformer is on or off. The foregoing described components in the heater control unit (17 ) serve to detect an adequate
threshold limit of current flowing through the lamps (12 ), and, in the presence of such threshold amount, turn the triac (69) off and thereby turn the heater
transformer off.
With continued reference to Fig. 5, the lamp starting unit (18) will now be described.
The output of the second operational amplifier (63 ) in the heater control unit (17) also connects
through a 10K ohm resistor (76) to the emitter of a 2N3904 transistor (77), the emitter of which also connects through a 1.8K ohm resistor (78) to the base thereof and to the 10 volt source (54). The base of this transistor also connects through a 1N4148 diode (79 ) to the collector thereof.
The collector of this transistor (77 ) also
connects through a bilateral trigger diode (81) (such as an HT32) to the gate of a silicon controlled
rectifier (B2) (such as an 560043F1). The anode
terminal of this silicon controlled rectifier (E2) connects through a 2 mega ohm one-half watt resistor (83 ) to the collector of the transistor (77), and also to a metal oxide varistor (84) (such as an ERZ C10 DK 621). The cathode terminal of the silicon controlled rectifier (82) connects to the low side of the power supply terminals (47), and the anode side connects to the disconnected lamp sensor and current limiter unit (19) as described below, and also through a .047 microfarad 600 volt capacitor (86) to the lamp unit (12 ).
In operation, the transistor (77) clamps the bilateral trigger diode (81 ) to a positive 10 volts when turned on. The resistor (76) connected to the heater control unit (17) provides the emitter current to turn the transistor (77) on. The bilateral trigger diode (81) provides gate exitation for the silicon controlled rectifier (82) and the capacitor (87) provides a charge for the trigger circuit. The varistor (84) serves to protect the silicon controlled rectifier (82) from over voltage transients and the silicon controlled rectifier (82) serves to pull down the common node between the two lamps in the lamp unit (12) to the low side of the voltage terminals (47) such that the first lamp (71) has the full starting voltage applied thereacross and the second lamp (72) is effectively cut off.
The current sensing circuitry of the heater control unit (17) will sense this condition, and provide an appropriate signal to cause the lamp starting unit (18) to shut the SCR (82) off, thereby applying the starting voltage across both lamps (71 and 72). Unless the lamps have both started by this point, the circuit control mechanism will reverse itself and the SCR (82) will again turn on and apply full voltage across only the first lamp (71 )• This process will continue in alternative fashion until all lamps in the lamp unit (12) are on.
With continued reference to Fig. 5, the
disconnected lamp sensor and current limiter unit (\9) will now be described. The disconnected lamp sensor and current limiter unit (19) includes a first and second sensing unit. The first sensing unit has a .22 ohm one watt resistor (88) connected between the low side of the power source terminals (47) and one secondary terminal of the heater transformer (70). One terminal of this resistor (88) also connects through a 10K ohm resistor (61e) to the noninverting input of an LM324 operational amplifier (89).
The output of this operational amplifier (89) connects to the inverting input thereof through a 330K ohm feedback resistor (91). The inverting input also connects through a 10K ohm resistor (61h) to the inverting input of a second LM324 operational
amplifier (92) through a 22 microfarad 6 volt
electrolytic integrating capacitor (93). The output of the first operational amplifier (89) also connects through two series connected 10K ohm resistors (61f and 61g ) to this same inverting input of the second operational amplifier (92), the noninverting input of which connects to positive 0.6 volt reference source (94

).
The output of this second operational amplifier (92 ) then connects through a 1K ohm resistor (96) to the light emitting diode side of an optical coupler (97) as provided through use of a 4N26. The emitter of the transistor side of this optical coupler (97) connects to the cathode side of the light emitting diode in a second optical coupler (98) as provided through use of an MOC 3012, and the collector of the transistor in the first optical coupler (97) connects through a 10K ohm resistor (99c) to the inverting input of an LM358 operational amplifier (101).
The second sensing unit has another current sensing .22 ohm one watt resistor (102) that connects at one terminal between the two lamps (7 1 and 72 ) and to ground. The high side of this resistor (loi)
connects through a 10K ohm resistor (99e) to the noninverting input of another LM358 operational amplifier (103) . The output of this operational
amplifier (103) connects through a 75K ohm feedback resistor (104) to the inverting input thereof, and also through a 10K ohm resistor (99d) to the collector of the transistor in the first optical coupler (97) and through a 10K ohm resistor (99c) mentioned above to the inverting input of the first LM358 operational amplifier (101) described above. The inverting input of this LM358 operational amplifier (ιol) also connects through a 10K ohm resistor (99b) to ground.
In addition, the first LM358 operational
amplifier (101) described above has its noninverting input connected through a diode (106) to ground and also through a 10K ohm resistor (99a) to a positive 5 volt source. The output of this operational amplifier (101) connects through a 62 ohm resistor (107) to the anode side of the light emitting diode of the second optical coupler (98). Finally, the inverting input of this operational amplifier (loi) connects through a 22 microfarad 6 volt electrolytic capacitor (108) to the collector of the transistor in the first optical coupler (97).
The second optical coupler (98) has a light sensitive triac having one terminal connected to the lamp unit (12) and the remaining terminal connected through a 33 ohm resistor (109) to the gate of a
Q6004F41 triac (111).
This latter triac (m) has an ERZ C10DK621 varistor (112) connected in parallel therewith, and one terminal connected to the H core dual transformer coupler of the ballast and phase differential control unit (16) as noted above and its remaining terminal connected to the light unit (12).
The first sensing unit resistor (88) serves to detect heater current with respect to the second lamp (72). The second sensing unit resistor (102) serves to detect heater current with respect to the first lamp (71 ). The associated operational amplifiers (89, 92, 103 and 101 ) serve to amplify and differentiate the sensed
signals in order to assess the presence or absence of heater current. The comparator outputs will be low when both lamps are installed and heater current is flowing. When a lamp has been removed, the optical couplers (97 and 98 ) will be activated and cause the center tapped mode of the H core dual transformer
coupler (20) of the ballast and phase differential control unit (16) to be connected across the light uirit (12) to thereby limit the available voltage from 250 volts AC RMS to 140 volts AC RMS. This in turn will limit available current to less than 5 milliamps, thereby substantially reducing the risk of accidental injury to maintenance personnel when manipulating
either lamp (7 1 or 72 ) .
To summarize, this ballast and control unit (10) provides a phase differential control unit (16) that presents a satisfactory power factor to the power
source and maintains output voltage and current in an appropriate relationship with one another to maximize output illumination. The heater control unit (17)
serves to sense when heater voltage must be applied during starting to aid in starting the flow of current through the electric discharge lamps. When this
process has begun, the heater voltage will be stopped. The lamp starting unit (18) alternately applies full available voltage across a single lamp, and then
across all lamps in series, until all lamps have been activated. Finally, the disconnected lamp sensor and current limiter unit (19) serves to detect the
presence of one or more disconnected lamps and, in response to this, limits available voltage and current to a safe limit.
Referring now to Fig. 4, a second (more
simplified) embodiment will be described.
Circuit (113) has a pair of input terminals (114) and (116) across which an AC voltage is impressed. A first inductor (117) is coupled to input terminal (114) and in series with a first capacitor (118). A bleed resistor (119) is across capacitor (118). A second inductor (121) is coupled to a point (122) common to first inductor (117) and first capacitor (118). The opposite end of the inductor (121) is ocupled to a lead (123). Thermal switch (114) is coupled between terminal (life) and point (126) on lead (123). Lead (123) is coupled to lamp (127), and capacitor (118) is coupled by a lead (128) to capacitor (129) and to lamp (131).
A transformer (102) has a primary winding coupled across leads (123) and (128) and has three secondary windings (133,134, and136) coupled to heater terminals of lamps (127 and 13 1) . Thus , heater voltages are supplied to the lamps when the primary winding of transformer (ill) is energized.
A second capacitor (129) is coupled in series with lead (128) and is coupled by a lead (137) to lead (138). Capacitor (129) is coupled by a lead (139) to one end of a voltage divider comprised of resistors (141,142, and 143. ). A third capacitor (144) is in a parallel with a resistor (141). A Triac (146) is between resistor (141) and lead (123 , resistor (142) being parallel to Triac (146 ) and Diac (147). A fourth capacitor (148) is in
parallel with resistor (143), capacitor (148) and
resistor (143) being also coupled to lead (123). A Diac (147) is coupled to the gate of Triac (146) and to the junction (149) between resistors (142) and (143).

First inductor (117) and first capacitor (118) define a series resonant circuit. A typical value for inductor (117) is two henries, and a typical value for capacitor (llδ) is two microfarads. Inductor (117) is typically a high Q inductor, and the values mentioned above are selected to form a tuned circuit at 80Hz with an input voltage of 277 volts RMS. Inductor (117) must be gapped to prevent core saturation.
Without inductor (121) in circuit (113), the series resonant circuit comprised of inductor (117) and capacitor (118) has a serious problem in that the high Q of this tuned circuit will cause a very large increase in voltage across the two circuit components, namely, inductor (117) and capacitor (118). Because the frequency of 80Hz is harmonically related to 60Hz, there is a very large increase of current through inductor (117), resulting in saturation of the core of this inductor. Thus, without inductor (121) in circuit (113), the circuit is not practical for power factor correction. With the use of inductor (121), the problems associated with power factor correction are eliminated.
Inductor (121), when it is in circuit (113) and the circuit is in operation, swamps out or eliminates the high Q of the series resonant circuit comprised of inductor (117) and capacitor (118). Moreover, the use of inductor (121) provides a storage element to improve lamp performance. If the value of inductor (121) is chosen to be 50 henries, the circuit Q will be dampened and inductor (117) will be prevented from saturating so as to prevent the Q of the circuit associated with inductor (117) from operating out of control.

A further refinement of circuit (113) is to provide for a common core for inductors (117 and 121 ). The performance of circuit (113) is also improved by adding mutual coupling between inductors (117) and (l21).
Capacitor (129) of circuit (113 ) (Fig.4) is used as a starting aid for lamp (127). However, to maintain a positive means for starting all lamps, the special circuit using Triac (146) and Diac (147) has been added to the ballast.
The operation of this special circuit is as follows: When the lamp (131) is not lit, the voltage at point (151) (Fig.4) will be 400 volts peak. This voltage level will, through resistive voltage divider, including resistors (141 ,142, and 143) , cause Diac (147) to trigger Triac (146). The Triac will turn on, and then back off again because the holding current requirement cannot be met. This action of the Triac will result in a voltage spike being developed and coupled through capacitor (144). This voltage spike will appear at lead (138), common to the two lamps and will force one of the lamps to start conduction.
When both lamps are lit, the voltage at point (151 ) will drop to 100 volts peak. This voltage level will cause Diac (147) to fail to reach turn-on level so as not to trigger Triac (146). This action assures that the starting circuit will be turned off whenever the lamps are lit. Resistor (141) can be of a very high value to minimize power wasted by the starting
circuit. Resistors (Wland 143) can be set to determine the voltage of the starting spike. Capacitor (148) is selected to place the turn on point of Triac (146) at the peak of the AC sign wave
Obviously, numerous modifications and variations of the present invention are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims, the invention may be practised otherwise than as specifically described therein.