Power Converter with Demand Pulse Isolation

CROSS-REFERENCE TO RELATED APPLICATIONS

This application Notice: The following multiple reissue applications have been filed for the reissue of U.S. Pat. No. 9,071,152: (1) reissue application Ser. No. 15/090,929, filed on Apr. 5, 2016 and reissued as U.S. Reissue Pat. No. RE47,031; (2) reissue application Ser. No. 15/168,998, filed on May 31, 2016 and reissued as U.S. Reissue Pat. No. RE47,713, which is a reissue continuation of Ser. No. 15/090,929; (3) reissue application Ser. No. 15/202,746, filed on Jul. 6, 2016 and reissued as U.S. Reissue Pat. No. RE47,714, which is a reissue continuation of Ser. No. 15/090,929; (4) the present reissue application Ser. No. 16/547,850, filed on Aug. 22, 2019, which is a reissue continuation of Ser. No. 15/202,746; (5) reissue application Ser. No. 16/548,897, filed on Aug. 23, 2019, which is a reissue continuation of Ser. No. 15/202,746; and (6) reissue application Ser. No. 16/987,654, filed on Aug. 7, 2020, which is a reissue continuation of Ser. No. 16/547,850. U.S. Pat. No. 9,071,152 claims the benefit of the filing dates of U.S. provisional application Nos. 61/667,473, filed on Jul. 03, 2012, and 61/727,795, filed on Nov. 19, 2012, the teachings of both of which are incorporated herein by reference in their entirety.

BACKGROUND

1. Field of the Invention

The present invention relates to electronics and, more specifically but not exclusively, to switched-mode power converters.

2. Description of the Related Art

This section introduces aspects that may help facilitate a better understanding of the invention. Accordingly, the statements of this section are to be read in this light and are not to be understood as admissions about what is prior art or what is not prior art.

Switched-mode DC-DC power converters, often powered by rectified DC from AC mains, are ubiquitous as plug-in adapters used to power a plethora of electronic devices.

A typical such converter is copiously documented in the Power Integrations Design example report DER-227. Such converters are also taught in U.S. Pat. No. 4,459,651 and U.S. Patent Application Publication Nos. 2011/0026277 A1 and 2011/0018590 A1. Such converters typically generate commutation pulses on the mains side of galvanic isolation circuitry.

Some known converters use forms of absorption modulation to convey feedback information through the power transformer. In U.S. Pat. No. 8,000,115, a temporary decrease in the loading of a transformer secondary winding during a flyback pulse generates a corresponding voltage disruption of the same pulse, which disruption is detected on another transformer winding to effect primary-winding-side converter control. In U.S. Pat. No. 5,973,945, a similar method is taught, but instead of unloading a flyback pulse, temporary loading of a forward power pulse is taught. The circuitry for extracting the resulting information-bearing current disruption in the transformer primary circuit is quite involved. A similar absorption modulator is taught in U.S. Pat. No. 4,996,638.

Converters are also known wherein an analog voltage reflection of the converter output voltage seen on a primary-side winding is processed to generate a primary-side analog feedback signal which is used to control the commutating signals applied to the commutating switch to regulate converter output on its secondary side. Such feedback methods are taught in U.S. Pat. Nos. 4,597,036 and 3,889,173. Such methods are becoming less common due to the difficulty of reliably processing the analog information reflected into a primary-side winding to obtain an accurate feedback signal.

U.S. Pat. No. 4,937,727 teaches a mains-side pulse generator that is pulse-width controlled by a voltage-responsive clamp on the output side of a galvanic isolation barrier.

BRIEF DESCRIPTION OF THE DRAWINGS

Other embodiments of the invention will become more fully apparent from the following detailed description, the appended claims, and the accompanying drawings in which like reference numerals identify similar or identical elements.

FIG. 1 shows a schematic diagram of a power converter according to an embodiment of the present invention using a blocking oscillator.

FIG. 2 shows a schematic diagram of a power converter according to another embodiment of the present invention using a blocking oscillator.

FIG. 3 shows a schematic diagram of a power converter according to an embodiment of the present invention using a simple transformer.

FIG. 4 shows a schematic diagram of a power converter according to an embodiment of the present invention using a separate pulse transformer.

DETAILED DESCRIPTION

FIG. 1 shows a schematic diagram of a power converter 10 a. A DC voltage source 5 a, external to this converter, which may be derived from AC mains, may be connected to an earth ground 6 a. Terminals 11 a and 12 a constitute a power input port that places source 5 a in circuit with a primary winding 101 a of a transformer 100 a and with a commutating switch 200 a, which is usually a MOSFET but may be a BJT or any other suitable electronic switch. For the diagrammed embodiment, switch 200 a is a MOSFET having a source S, a gate G, and a drain D. Transformer 100 a also comprises a regeneration winding 102 a which is referenced to source S of MOSFET 200 a, is connected through a capacitor 202 a to gate G of MOSFET 200 a, and is poled to provide regenerative feedback to gate G of MOSFET 200 a. Connected between terminal 11 a and gate G of MOSFET 200 a is a resistor 201 a which charges capacitor 202 a to enhance MOSFET 200 a at a slow pulse rate. Thus, MOSFET 200 a, transformer 100 a, capacitor 202 a, and resistor 201 a form an input-side blocking oscillator which acts as a driver circuit toggling ON and OFF MOSFET 200 a.

Transformer 100 a also comprises a secondary winding 104 a which may be connected to a floating common terminal 14 a. A diode 300 a and a capacitor 301 a form a rectifier circuit to rectify and filter voltage pulses from winding 104 a to supply power through a power output port comprising terminals 13 a and 14 a to an external load represented by resistor 7 a connected in circuit therewith, one end of which may be referred to a floating common 8 a. The power input port 11 a/ 12 a and the power output port 13 a/ 14 a may be galvanically isolated from each other.

Flyback pulses of transformer 100 a occur when MOSFET 200 a ceases conduction, i.e., turns OFF. Winding 104 a is poled to cause diode 300 a to rectify only these flyback pulses.

Forward pulses, of opposite polarity to the flyback pulses, occur while MOSFET 200 a is ON. Another diode 500 a, poled to rectify forward pulses, and another capacitor 501 a form an auxiliary rectifier circuit to rectify and filter forward pulses from winding 104 a, and to store energy for triggering the input-side blocking oscillator formed by MOSFET 200 a, transformer 100 a, capacitor 202 a, and resistor 201 a. Resistor 201 a is made sufficiently large to set a low free-running frequency of the blocking oscillator, perhaps 1 KHz or less, to minimize power consumption. Nevertheless, the miniscule power thus provided suffices to charge capacitor 501 a to a voltage related, through the turns ratio of transformer 100 a, to the voltage at the power input port, even with the power output port short-circuited.

This magnetically-coupled blocking oscillator may be triggered through any transformer winding magnetically coupled thereto. Therefore, just as MOSFET 200 a may be turned ON through winding 102 a, it may as easily be triggered through winding 104 a. To trigger thusly, diode 500 a is briefly short-circuited by a switch 502 a which is driven by a demand pulse generator 503 a to source a pulse of energy from capacitor 501 a into transformer 100 a. When this is done, the voltage at the cathode of diode 500 a falls rapidly to the voltage on its anode, also being the voltage across capacitor 501 a. Since winding 104 a is coupled to winding 101 a, the voltage on drain D of MOSFET 200 a also rapidly falls from near the voltage on terminal 11 a to near the voltage on terminal 12 a. Since winding 102 a is also magnetically coupled, the voltage at its node shared with capacitor 202 a abruptly rises, turning ON MOSFET 200 a. This triggering action occurs in a few tens of nanoseconds. Until regeneration is established in MOSFET 200 a through winding 102 a, triggering energy is supplied by capacitor 501 a of the auxiliary rectifier circuit. However, once regeneration is established in MOSFET 200 a, capacitor 501 a is charged for the duration of the ON time of MOSFET 200 a, fully replacing any energy lost during triggering. The demand pulse generator 503 a may be used to adjust the commutation frequency of the converter 10 a to cause its output to attain a desired value, as will be described below.

It is important to understand certain important advantages of this embodiment. Firstly, this embodiment allows minimal, simple, and robust circuitry to be galvanically associated with the power input port where high voltages and mains transients may be expected. According to this embodiment, more complex and vulnerable regulation circuitry may be galvanically associated with the power output port where voltages are often lower and protection is more easily implemented. Secondly, the control of a flyback converter that may cross from the discontinuous conduction mode (DCM), through the critical conduction mode, to the continuous conduction mode (CCM), is well known to be problematic. This embodiment simply avoids that problem. In the embodiment shown in FIG. 1 , transformer 100 a is used during the conduction of MOSFET 200 a as a forward converter supplying the auxiliary rectifier circuit, and during the flyback of transformer 100 a as a flyback converter supplying power to the power output port. During these cycle portions, it is difficult and impractical to re-trigger the blocking oscillator through transformer 100 a to generate another energy-bearing cycle. Once the flyback pulse has reset the inductance of transformer 100 a, i.e., has depleted energy from its magnetic field, transformer 100 a is free, until the next ON time of MOSFET 200 a, to be used as a magnetically coupled isolator to convey trigger information between its windings. In FIG. 1 , the information thus conveyed is a pulse from pulse generator 503 a which, responsive to the output of comparator 401 a, indicates the need for another energy-bearing cycle, and moreover re-triggers the blocking oscillator to provide that energy-bearing cycle. Since it is difficult or impractical to re-trigger until transformer 100 a energy has been depleted, this converter will, if driven as hard as possible, approach critical conduction, but refuse to enter the critical conduction mode.

This converter may be fitted with a reference voltage 400 a and a comparison circuit 401 a. When the voltage at terminal 13 a falls below the comparison voltage, comparison circuit 401 a causes pulse generator circuit 503 a to pulse, turning ON switch 502 a, triggering an energy-bearing ON cycle of the blocking oscillator, and charging capacitor 301 a. As load 7 a drains capacitor 301 a, terminal 13 a voltage repeatedly falls to the voltage of reference 400 a, causing comparison circuit 401 a to initiate energy-bearing ON cycles. An interesting property of this embodiment is that the bottom of its output ripple corresponds to the voltage of reference 400 a, and the amplitude of its ripple decreases with increased current in load 7 a.

FIG. 2 shows a schematic diagram of a power converter 10 b. As in converter 10 a of FIG. 1 above, converter 10 b is powered, through terminals 11 b and 12 b, from an external source 5 b, that may be referred to earth ground 6 b. Power from converter 10 b flows through terminals 13 b and 14 b through a load 7 b, which may be referred to a floating common 8 b. A MOSFET 200 b, preferably ON Semiconductor type NDD02N60, forms an input-side blocking oscillator with a (preferably 1 nF) capacitor 202 b, a (preferably 66 megohm) resistor 201 b, and a transformer 100 b. Transformer 100 b comprises a winding 101 b, preferably about 250 uH, and windings 102 b and 104 b, preferably about 3.09 uH each, and a winding 103 b, preferably about 193 nH, which may be a single turn. A capacitor 212 b provides a short local circuit for high frequency currents and preferably comprises a 4.7 uF capacitor and a 100 nF capacitor (neither explicitly shown) in parallel. A resistor 210 b, preferably about 180 ohms, and a capacitor 211 b, preferably about 10 pF, filter out capacitive spikes generated by fast transitions of MOSFET 200 b. When MOSFET 200 b is turned on, the current therein rises, but is limited by a transistor 208 b, the base of which is driven by a voltage across a resistor 209 b, which voltage is responsive to current through MOSFET 200 b. When MOSFET 200 b current reaches about 250 mA, transistor 208 b shunts current at gate G of MOSFET 200 b to ground, limiting gate G voltage to prevent further current rise. With current rise prevented, the voltages across the windings of transformer 100 b collapse. Thus, a regenerative turn-OFF of MOSFET 200 b begins, and the voltage at its drain D flies positive past the voltage on terminal 11 b until the energy in its magnetic field finds a current path through one of its windings. A corresponding negative voltage occurs at the shared node of winding 102 b and a capacitor 204 b, preferably about 100 pF, which immediately couples through a resistor 207 b, preferably about 47 ohms, vigorously turning off MOSFET 200 b. Within a few nanoseconds, the same transition couples through a resistor 203 b, preferably about 1K, and a capacitor 202 b, preferably about 1 nF, to join the signal passing through capacitor 204 b, to reinforce the OFF transition at gate G of MOSFET 200 b. Both the OFF and ON transitions at gate G of MOSFET 200 b are regenerative and follow the path just described. To prevent damage to MOSFET 200 b, its gate voltage should be limited. Resistor 203 b and a diode 205 b, preferably an 8.2 volt zener, form an L-network to limit that voltage. Since the voltage at the cathode of diode 205 b is capacitively coupled to resistor 207 b, and resistor 201 b is pulling up on resistor 207 b, the gate G voltage of MOSFET 200 b would be free to rise, turn ON MOSFET 200 b continually, and perhaps damage its gate, if means for limiting gate voltage were not provided. Another zener diode 206 b, preferably the zenered base-emitter junction of an NXP type PMBT 3904, is used to limit the gate voltage rise. This device is used because, at high temperature, excess leakage of diode 205 b would shunt to ground the current of resistor 201 b, preventing the blocking oscillator from starting. Most of the current from winding 102 b, being too great for diode 206 b to conduct without damage, flows in diode 205 b.

As in FIG. 1 , when the voltage at the node of winding 104 b and a diode 300 b, preferably type 1N4148, flies back, the energy in transformer 100 b is dumped into a capacitor 301 b, preferably 4.7 uF, ultimately to be consumed by load 7 b. As in FIG. 1 , a diode 500 b, preferably type 1N4148, and a capacitor 501 b, preferably 220 nF, form a forward converter to supply an auxiliary voltage.

Please note that, in this embodiment, the poling of winding 104 b, diode 300 b, and diode 500 b are reversed, and the output polarity is reversed, with respect to FIG. 1 . This reversal illustrates that this embodiment will function with either poling, and that polarity is of little practical concern in an isolated supply. The rectifiers and windings are so poled that the auxiliary supply forms a forward converter, and the output forms a flyback converter with the remaining circuitry. A switch 502 b is, in this embodiment, a PNP transistor, preferably type MMBT 3906. This switch also coacts with another winding 103 b of transformer 100 b, which may be a single turn, and a capacitor 504 b, to form an output-side triggering blocking oscillator 503 b corresponding to pulse generator 503 a of FIG. 1 , which triggering blocking oscillator is magnetically coupled through transformer 100 b to the above-described input-side, power-blocking oscillator comprising MOSFET 200 b. Thus, an input-side, master blocking oscillator comprising MOSFET 200 b and an output-side, slave blocking oscillator comprising switch 502 b are magnetically coupled to each other through transformer 100 b. The auxiliary voltage of capacitor 501 b flows through a resistor 406 b, preferably 27K, to feed another zener diode 400 b, preferably another zenered PMBT3904, corresponding to reference 400 a of FIG. 1 . A capacitor 407 b, preferably 100 nF, bypasses diode 400 b at high frequencies. A dual transistor 402 b, preferably NXP type BS846, is connected as a current mirror, and mirrors the current in a resistor 404 b, the latter current being set by the reference voltage of diode 400 b. This current sets the free running oscillation frequency of blocking oscillator/pulse generator 503 b. Since the transformer 100 b current flowing in MOSFET 200 b is set by transistor 208 b, the per-cycle energy in transformer 100 b is quantized. Setting the current in resistor 404 b, preferably 100K, therefore sets a maximum frequency of blocking oscillator/demand-pulse generator 503 b, thereby setting the maximum frequency for these energy-quantized cycles, thus limiting maximum converter power, even in the event of an output short-circuit. A diode 403 b, preferably type 1N4148, compensates the base-emitter voltage of dual-transistor 402 b. Dual transistor 402 b, resistor 404 b, diode 403 b, and a resistor 405 b form a current comparator corresponding to comparator 401 a of FIG. 1 . Resistor 405 b, preferably about 82K for a 5V output, provides feedback by robbing resistor 404 b current from the current mirror of dual-transistor 402 b as output voltage increases, thus setting operating frequency roughly in proportion to the demand of load 7 b.

FIG. 3 shows a schematic diagram of a power converter 10 c arranged to use a simple, two-winding transformer. The function of converter 10 c closely parallels that of converters 10 a and 10 b of FIGS. 1 and 2 , save that components have been added to replace the functions of regenerative (tickler) windings needed by blocking oscillators. As in the previous figures, a source 5 c may be referenced to an earth ground 6 c, and load 7 c may be referenced to a floating common 8 c. As in FIG. 1 , a transformer 100 c primary winding 101 c is in circuit with a switch 200 c and terminals 11 c and 12 c. As in FIG. 1 , flyback pulses on a secondary winding 104 c of transformer 100 c charge a capacitor 301 c through a diode 300 c to supply energy to the load 7 c through terminals 13 c and 14 c. As in FIG. 1 , forward pulses on the secondary winding of transformer 100 c charge a capacitor 501 c through a diode 500 c. As in FIG. 1 , a comparison circuit 401 c compares the voltage on the output terminal 13 c with a reference 400 c. As in FIG. 1 , a switch 502 c is driven by a demand pulse generator 503 c.

We now depart from the FIGS. 1 and 2 function. A fast oscillator 505 c, preferably about 100 KHz, drives an AND gate 506 c which is also driven by comparison circuit 401 c. If the voltage between terminals 13 c and 14 c is smaller than that of reference 400 c, gate 506 c passes oscillator 505 c pulses to trigger pulse generator 503 c, which initiates, through transformer 100 c additional energy-bearing pulses by eventually driving switch 200 c, as described below. If, however, the output voltage is adequate, then gate 506 c does not pass oscillator 505 c pulses.

The pulses of energy from capacitor 501 c sourced to transformer 100 c, though a switch 502 c, during the pulse of generator 503 c, under the command of gate 506 c, appear as voltage pulses across the primary winding of transformer 100 c. These pulses are detected and processed to logic levels by a demand pulse detector 215 c and passed through an OR gate 214 c to a pulse generator that turns ON switch 200 c to energize transformer 100 c to begin an energy-bearing cycle. When switch 200 c turns OFF, the subsequent flyback pulse charges capacitor 301 c through diode 300 c, as previously described. Since capacitor 501 c is charged from the converter forward pulse, its voltage persists even in the presence of a short-circuit load, allowing the converter to recover once the short-circuit is removed.

Had no energy-bearing cycle ever occurred, there might be insufficient, or no, charge in capacitor 501 c to be used to initiate energy-bearing cycles as described above. Therefore, a slow pulse oscillator 213 c, preferably about 1 KHz, is also connected to gate 214 c, through which it initiates energy-bearing cycles by triggering a pulse generator 216 c, thus turning on switch 200 c. These infrequent pulses cause energy-bearing cycles that are sufficient to charge capacitor 501 c, which also may supply power to generator 503 c, gate 506 c, oscillator 505 c, reference 400 c, and comparison circuit 401 c. Of course, slow oscillator 213 c must somehow be powered along with gate 214 c and pulse generator 216 c. A bias supply (not shown but well known in the art) powered from terminals 11 c and 12 c, may be used to power these components of the circuit.

FIG. 4 shows a schematic diagram of a power converter 10 d, comprising a separate transformer 110 d to transmit demand pulses across a galvanic isolation barrier. As in FIG. 1 above, converter 10 d is powered, through terminals 11 d and 12 d, from an external source 5 d, and power output from converter 10 d flows through terminals 13 d and 14 d.

Input voltage from terminals 11 d and 12 d powers a slow oscillator 213 d, preferably of less than 1 KHz frequency, and a start-up regulator 232 d which, through a supply node + 5 d, initially powers, with a voltage preferably about 4V, logic and drive circuitry described below. Each label “+ 5 d” in FIG. 4 refers to a supply node that is initially about 4 volts when the input-side logic is starting to function and about 5 volts when in regulation. A capacitor 221 d and a resistor 222 d differentiate transitions of a slow pulse oscillator 213 d to provide pulses of about 200 nS duration. These pulses pass though a NAND gate 223 d to clock a D-type flip-flop 220 d through a node CKa.

Responsive to its clock pulse, flip-flop 220 d turns ON a switch 200 d, preferably a MOSFET, ON Semiconductor type NDD02N60, which is in circuit with a primary winding 101 d of a transformer 100 d, with a sense resistor 209 d, and with terminals 11 d and 12 d. Current then flows in this circuit, and the voltage of source 5 d is impressed upon primary winding 101 d. According to the turns-ratio between primary winding 101 d and a secondary winding 104 d of transformer 100 d, a voltage appears across winding 104 d. This latter voltage charges a capacitor 416 d through a diode 417 d.

As current in resistor 209 d rises, a voltage is applied to an input of a comparator 217 d, which voltage is compared with a reference 216 d, also connected to an input of comparator 217 d. When current in resistor 209 d exceeds a value set by reference 216 d, comparator 217 d issues a reset signal which propagates through NAND gates 218 d and 219 d to a node /Ra where the reset signal resets flip-flop 220 d, turning OFF switch 200 d.

When switch 200 d is turned ON, unavoidable gate-to-source capacitance of MOSFET switch 200 d causes a current spike in resistor 209 d. To prevent comparator 217 d from prematurely resetting flip-flop 220 d responsive to this spike, the rise of node Qa charges a capacitor 231 d through a resistor 230 d to reach the threshold of a gate 219 d in about 75 nS, prior to which the low voltage of capacitor 231 d inhibits gate 219 d from resetting flip-flop 220 d.

Prior to its rise, node Qa has been low, and a complementary node/Qa has been high. When node Qa rises, node/Qa falls, discharging a capacitor 229 d through a resistor 228 d to the threshold of NAND gate 218 d in about 2 uS, and though NAND gate 219 d resetting flip-flop 220 d, thus limiting the maximum ON time of switch 200 d, should comparator 217 d fail to reset flip-flop 220 d.

In addition to limiting ON times of switch 200 d, it is desirable to limit maximum frequency of these ON times. To this end, the voltage across a capacitor 226 d is charged to a logic high through a resistor 225 d and applied to a node Da, the D-input of flip-flop 220 d. When node/Qa falls, capacitor 226 d is discharged through a diode 227 d, slowly to be recharged through resistor 225 d. Until the capacitor 226 d voltage is recharged to the D-input threshold voltage, flip-flop 220 d is inhibited from turning ON switch 200 d.

When switch 200 d is turned OFF, the energy in the magnetic field of transformer 100 d generates flyback voltage across its windings. Flyback voltage of winding 104 d is rectified by a diode 300 d and begins to charge a filter capacitor 301 d to begin to supply output voltage to terminals 13 d and 14 d. This flyback voltage also raises the voltage on capacitor 416 d, causing diode 417 d to turn OFF and a diode 418 d to turn ON, charging a capacitor 419 d. Voltage across capacitor 419 d supplies an auxiliary regulator 420 d, which in turn powers a fast oscillator 505 d, preferably of about 60 KHz frequency. Regulator 420 d also powers logic and drive circuitry on the winding 104 d side of the power converter.

The ON pulses of switch 200 d responsive to oscillator 213 d are sufficiently frequent to start the converter of this embodiment, but insufficiently frequent to drive it to full output. To initiate more frequent pulses, an oscillator 505 d drives a capacitor 507 d and a resistor 508 d to supply differentiated pulses of about 100 nS width to a NAND gate 509 d, which in turn drives a primary winding 111 d of demand pulse transformer 110 d, thus producing demand pulses across a secondary winding 112 d thereof. These winding 112 d pulses are conveyed through a NAND gate 223 d to clock flip-flop 220 d at up to the frequency of oscillator 505 d.

If all of the pulses of oscillator 505 d were allowed to clock flip-flop 220 d, under some conditions, the converter of this embodiment would produce excess output. To regulate this output, a flip-flop 412 d is used to gate the pulses passed by NAND gate 509 d. At a node CKc, oscillator 505 d clocks a flip-flop 412 d, which generates a logic high at a node Qc only when a logic high is present at a node Dc at the rising edge of its clock. Thus, pulses driving transformer 110 d are permitted responsive to a logic high only at node Dc.

It would be wasteful of power to drive winding 111 d for the full duration of the differentiated pulse at resistor 508 d. Therefore, when switch 200 d turns ON causing a negative transition at the dotted end of winding 101 d, a corresponding negative transition appears at the dotted end of winding 104 d. This transition is coupled through a small capacitor 414 d, preferably about 10 pF, through a current-limiting resistor 415 d to a node/Rc, the reset input of flip-flop 412 d, which is normally held high by a resistor 413 d. Thus, once the turning ON of switch 200 d has propagated through transformer 100 d, flip-flop 412 d is reset, usually in less than 20 nS.

Node Dc is usually held at a logic high by a resistor 411 d, thus enabling pulses gated by flip-flop 412 d. However, between terminals 13 d and 14 d is disposed a voltage divider comprising resistors 408 d and 409 d, the voltage at the junction of which is applied to an input of a comparator 401 d. Should the voltage at that junction exceed the voltage of a reference 400 d, also applied to a comparator 401 d input, an output of comparator 401 d will drop to a logic low, drawing current through a diode 410 d, thus presenting a logic low at node Dc and, after clocking, responsively at node Qc, inhibiting pulses through gate 509 d that would otherwise turn ON switch 200 d. Thus, the voltage between terminals 13 d and 14 d is regulated responsive to the voltage of reference 400 d.

Since the voltage between terminals 11 d and 12 d may be high, perhaps 375V, and the desired regulated voltage at node + 5 d is typically 5V, it might be inefficient to obtain the power to supply the logic and drive circuitry associated with winding 101 d from regulator 232 d. Therefore, transformer 100 d is fitted with an auxiliary winding 102 d, which is connected in circuit with an inductor 235 d, a diode 241 d, and a switch 233 d, preferably a MOSFET. While switch 200 d is ON, current flows in this circuit. When switch 200 d turns OFF, diode 241 d also turns OFF and energy in inductor 235 d generates a positive flyback voltage, causing current through a diode 236 d to charge a filter capacitor 237 d, raising the voltage of node + 5 d. As node + 5 d approaches 5V, regulator 232 d ceases to supply energy to node + 5 d, but continues to power a voltage reference 242 d, which drives an input of a comparator 240 d. Should the voltage of node + 5 d exceed 5V, the voltage at the junction of resistors 238 d and 239 d, connected to another input of comparator 240 d, will exceed that of reference 242 d, causing the output of comparator 240 d at node Db to drop to a logic low.

A flip-flop 234 d drives node Qb to turn ON switch 233 d responsive to clock pulses on node Qa, and to a logic high being present at node Db. When node Db drops to a logic low, node Qb follows it upon the next clock, and switch 233 d turns OFF. In this state, inductor 235 d no longer receives energy and no longer charges capacitor 237 d through diode 236 d. Thus, node + 5 d is regulated to approximately 5V, and the energy supplying node + 5 d is provided efficiently through transformer 100 d.

In one embodiment, the invention is a switched-mode power-converter comprising a power input port, a transformer comprising windings, a commutating switch connected in circuit with the input port and a winding of the transformer, a driver circuit for toggling the commutating switch, a power output port, a rectifier circuit for supplying power to the power output port, a reference voltage or current source, a comparison circuit for comparing the voltage or current at the power output port with the reference voltage or current, and a demand pulse source circuit coupled to the transformer for transmitting galvanically isolated trigger information through the transformer to the driver circuit responsive to the comparison circuit.

The converter may comprise as its driver circuit a blocking oscillator comprising the converter transformer. The converter may further comprise an input-side, master blocking oscillator for power conversion and an output-side, slave blocking oscillator for generating demand pulses. Both blocking oscillators may be mutually coupled through the converter power transformer or may drive separate transformers.

The converter may comprise inductive, capacitive, opto-coupled, or piezoelectric galvanic isolation circuitry to transmit demand pulses across the galvanic isolation barrier.

The converter may have one or more output rectifier circuits poled to rectify flyback pulses of its transformer.

The converter may comprise one or more auxiliary rectifier circuits which may be poled as forward converters.

The converter may be powered by a rectifier circuit to provide an AC/DC converter.

It should be understood that replicas of pulses generated and applied to one winding of the power transformer appear, suitably modified by turns-ratio, across all other windings of the power transformer.

Though blocking oscillators usually require tickler windings, single output embodiments of this invention may comprise a power transformer with as few as two, and in excess of five windings, with multiple output embodiments possibly comprising yet more windings.

Startup pulse generation circuitry resides on the powered side of the isolation barrier, though its pulses appear on both sides of the isolation barrier. This circuitry may comprise a blocking oscillator, another form of oscillator with drive circuitry to turn ON the commutating switch, or this circuitry may comprise an external source of pulses.

Demand pulse generator circuitry resides with the output port to be regulated, though its pulses appear on both sides of the isolation barrier. This circuitry may comprise a slave blocking oscillator, another form of oscillator with drive circuitry to turn ON the demand pulse generator switch, or may be externally applied.

Demand pulses may be generated to regulate the power converter to provide either a desired output voltage or a desired output current responsive to the voltage across or a current through an output port.

Between the commencement of start-up and the attainment of regulation, a pulse generator sources pulses to turn ON the commutating switch. This pulse generator may be the same generator that sources regulation pulses, or may be a separate pulse generator.

Each internal pulse generator is powered. The startup pulse generator is powered from the input port. The demand pulse generator is only indirectly powered from the input port by DC-DC power conversion through the power transformer and one or more rectifiers and filters powering the power output port with which the generator is associated.

Power for pulse generation circuitry may be rectified from either forward pulses, from flyback pulses, or both, appearing across one or more power transformer windings. Rectification of forward pulses helps to assure startup.

Windings, switches, and diodes may be poled to provide either polarity of input, and either polarity of output.

In each of the embodiments of FIGS. 1 - 4 , galvanic isolation circuitry transfers (i) power from the input-port side to the output-port side of the power converter and (ii) demand pulses from the output-port side to the input-port side. In particular, in FIG. 1 , the galvanic isolation circuitry consists of transformer 100 a, which transfers (i) power from winding 101 a to winding 104 a and (ii) demand pulses from winding 104 a to winding 102 a. In FIG. 2 , the galvanic isolation circuitry consists of transformer 100 b, which transfers (i) power from winding 101 b to winding 104 b and (ii) demand pulses from winding 104 b to winding 102 b. In FIG. 3 , the galvanic isolation circuitry consists of transformer 100 c, which transfers (i) power from winding 101 c to winding 104 c and (ii) demand pulses from winding 104 c to winding 101 c. In FIG. 4 , the galvanic isolation circuitry consists of (i) transformer 100 d, which transfers power from winding 101 d to winding 104 d and (ii) transformer 110 d, which transfers demand pulses from winding 111 d to winding 112 d. Winding 102 d generates the bias supply for powering the + 5 d node.

In each of the embodiments of FIGS. 1 - 4 , a demand pulse generator on the output-port side of the converter generates the demand pulses that are conveyed to the input-port side of the converter via the galvanic isolation circuitry. In FIGS. 1 , 2 , and 3 , the demand pulse generator comprises elements 503 a, 503 b, and 503 c, respectively. In FIG. 4 , the demand pulse generator comprises NAND Gate 509 d and flip-flop 412 d.

In each of the embodiments of FIGS. 1 - 4 , slow-pulse source circuitry generates pulses on the input side of the power converter. In FIGS. 1 and 2 , the slow-pulse source circuitry is the corresponding input-side blocking oscillator. In FIGS. 3 and 4 , the slow-pulse source circuitry is slow oscillator 213 c and slow oscillator 213 d, respectively. Note that, depending on the particular implementation, the slow-pulse source circuitry may be implemented internal to or external to the switched-mode power converter. Similarly, depending on the particular implementation, fast oscillator 505 c and fast oscillator 505 d of FIGS. 3 and 4 , respectively, may be implemented internal to or external to the switched-mode power converter.

Embodiments of the invention may be implemented as (analog, digital, or a hybrid of both analog and digital) circuit-based processes, including possible implementation as one or more integrated circuits (such as an ASIC or an FPGA), a multichip module, a single card, or a multicard circuit pack.

Also for purposes of this description, the terms “couple,” “coupling,” “coupled,” “connect,” “connecting,” or “connected” refer to any manner known in the art or later developed in which energy or signals are allowed to be transferred between two or more elements, and the interposition of one or more additional elements is contemplated, although not required. Conversely, the terms “directly coupled,” “directly connected,” etc., imply the absence of such additional elements.

Also, for purposes of this disclosure, it is understood that all gates are powered from a fixed voltage power domain (or domains) and ground unless shown otherwise. Accordingly, all digital signals generally have voltages that range from approximately ground potential to that of one of the power domains and transition (slew) quickly. However and unless stated otherwise, ground may be considered a power source having a voltage of approximately zero volts, and a power source having any desired voltage may be substituted for ground. Therefore, all gates may be powered by at least two power sources, with the attendant digital signals therefrom having voltages that range between the approximate voltages of the power sources.

Signals and corresponding nodes or ports may be referred to by the same name and are interchangeable for purposes here.

Transistors are typically shown as single devices for illustrative purposes. However, it is understood by those with skill in the art that transistors will have various sizes (e.g., gate width and length) and characteristics (e.g., threshold voltage, gain, etc.) and may consist of multiple transistors coupled in parallel to get desired electrical characteristics from the combination. Further, the illustrated transistors may be composite transistors.

The terms “source,” “drain,” and “gate” should be understood to refer either to the source, drain, and gate of a MOSFET or to the emitter, collector, and base of a bipolar device when an embodiment of the invention is implemented using bi-polar transistor technology. p Unless explicitly stated otherwise, each numerical value and range should be interpreted as being approximate as if the word “about” or “approximately” preceded the value of the value or range.

It will be further understood that various changes in the details, materials, and arrangements of the parts which have been described and illustrated in order to explain embodiments of this invention may be made by those skilled in the art without departing from embodiments of the invention encompassed by the following claims.

The use of figure numbers and/or figure reference labels in the claims is intended to identify one or more possible embodiments of the claimed subject matter in order to facilitate the interpretation of the claims. Such use is not to be construed as necessarily limiting the scope of those claims to the embodiments shown in the corresponding figures.

It should be understood that the steps of the exemplary methods set forth herein are not necessarily required to be performed in the order described, and the order of the steps of such methods should be understood to be merely exemplary. Likewise, additional steps may be included in such methods, and certain steps may be omitted or combined, in methods consistent with various embodiments of the invention.

Although the elements in the following method claims, if any, are recited in a particular sequence with corresponding labeling, unless the claim recitations otherwise imply a particular sequence for implementing some or all of those elements, those elements are not necessarily intended to be limited to being implemented in that particular sequence.

Reference herein to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments necessarily mutually exclusive of other embodiments. The same applies to the term “implementation.”

The embodiments covered by the claims in this application are limited to embodiments that (1) are enabled by this specification and (2) correspond to statutory subject matter. Non-enabled embodiments and embodiments that correspond to nonstatutory subject matter are explicitly disclaimed even if they fall within the scope of the claims.