Transformers are often used to transfer information or power between circuits that are operating at different voltages or under different noise conditions. In many circuit arrangements, a logic signal must be transmitted between two circuits that must otherwise be electrically isolated from one another. For example, the transmitting circuit could utilize high internal voltages that would present a hazard to the receiving circuit or individuals in contact with that circuit. In the more general case, the isolating circuit must provide both voltage and noise isolation across an insulating barrier.
One type of galvanic isolator utilizes a transformer based system to isolate the two circuits. The sending circuit is connected to the primary coil of the transformer and the receiving circuit is connected to the secondary coil. The information is transferred by modulating the magnetic field generated in the primary coil. In this arrangement, the sending and receiving circuits can utilize entirely different power supplies and grounds and operate at different signal voltage levels. Typically, the transmitter and the two windings are constructed on a first semiconductor chip and the receiver is constructed on a separate chip that is connected to the first chip by wire bonds or the like. The two transformer windings are, typically, deposited over or near the drive circuits on the first chip by patterning two of the metal layers that are typically provided in conventional semiconductor fabrication processes. Alternatively, the coils may be fabricated on a different chip.
If the transformer coils are fabricated on the transmitter chip, the size of the transmitter chip is set by the size of the transformer coils, which typically require a significant area of silicon compared to the drive circuitry. Alternatively, if the coils are fabricated on the receiver chip or a separate chip, the coils will still require a significant area of silicon on those chips. The cost of the semiconductor substrate is a significant fraction of the cost of the isolator. This is a particularly significant problem when large coils are required to provide the coupling between the transmitter and receiver. In addition, many applications require multiple independent galvanic isolators on a single substrate. Cross-talk between the isolators constructed on silicon substrates using conventional semiconductor fabrication techniques is difficult to block in a cost-effective manner because of fringe fields generated by one coil being coupled to an adjacent coil. If the chips are separated by a sufficient distance on the silicon substrate, the cost of the wasted silicon becomes significant.
In addition to the wasted silicon area, devices constructed using conventional silicon integrated circuit fabrication have limitations that are imposed by the design rules of the fabrication line and the limitations as to materials that are allowed on that line. For many applications, the dielectric insulation between the coils of the transformer must withstand voltages in excess of 1000 volts. The thickness of dielectric that is available in conventional CMOS fabrication lines is insufficient to provide this degree of insulation. In addition, in some applications it would be advantageous to provide a ferrite layer or layers near the coils of the transformer to improve the coupling efficiency. However, the materials in question cannot be utilized in many conventional fabrication lines.
In some cases, it would be advantageous to power one of the circuits from the other circuit. For example, the transmitting circuit could power the receiving circuit. Such an arrangement would allow the receiving circuit to operate at different voltages than the transmitting circuit without requiring a separate power source on the receiving circuit. In principle, a transformer could also be utilized to provide the power transfer function. However, the efficiency required to provide the power transfer function is significantly greater than that needed to merely transmit information. Hence, such transformers are not easily, or economically, constructed using silicon-based fabrication techniques.
Miniature transformers constructed by winding wire around small cores are also known to the art. However, these devices are made one at a time, and hence, lack the economies of scale that are provided by wafer-scale photolithographic techniques and other mass production techniques developed for integrated circuits and the packaging thereof. Miniature transformers made by plating the coil pattern for the primary coil winding on one side of a printed circuit board and the secondary winding on the other side of the printed circuit board are also known. However, these dielectric core transformers have insufficient windings and are required to operate at relatively high frequencies because of the lack of a soft ferrite core.
The present invention includes a component coil for constructing transformers and the transformer constructed therefrom. A component coil according to the present invention includes a substrate having an insulating layer of material having top and bottom surfaces. The top surface includes a first trace having an outer end and an inner end and a first spiral conductor connected between the outer and inner ends of the first trace. The bottom surface includes a second trace having an outer end and an inner end and a second spiral conductor connected between the outer and inner ends of the second trace. A conductor connects the inner ends of the first and second traces. The outer ends of the first and second traces are connected to first and second contacts, respectively. The first and second spiral conductors are oriented such that a current traveling from the outer end of the first trace to the inner end of the first trace generates a magnetic field having a first component perpendicular to the top surface, and a current passing from the inner end of the second trace to the outer end of the second trace generates a magnetic field having a second component perpendicular to the top surface. The first component has a direction that is the same as the second component.
A transformer according to the present invention includes a primary winding and a secondary winding in which one of the windings is a first component coil. An insulator separates the primary and secondary windings. The first component coil is aligned with the other of the primary and secondary windings such that a portion of the magnetic field generated by the first component coil passes through the other winding when a potential difference is applied between power pads of the first component coil. In one aspect of the invention, the other of the primary and secondary windings includes a second component coil and the primary or secondary winding includes a third component coil aligned with the first component coil such that a portion of the magnetic field generated by the third component coil passes through the first trace in the second component coil when a potential difference is applied between the power pads of the first component coil, or second component coil, respectively. In another aspect of the invention, the first component coil includes a layer of magnetically-active material.
A transformer according to the present invention is constructed by combining a number of component coils to form the primary and secondary windings of the transformer. Each component coil is constructed on an insulating substrate and includes first and second traces that can be generated using conventional photolithographic techniques of the type utilized in making printed circuit boards or semiconductor devices.
The manner in which the present invention provides its advantages can be more easily understood with reference to
The portions of the traces that are designed to generate the magnetic fields that couple the various windings in transformers constructed from the component coils are topologically spirals. While the drawings show generally circular spirals, any linear pattern that winds in a continuous and gradually widening curve around a central region can be utilized. The spirals are configured such that a current flowing through one of the spirals generates a magnetic field with a component that is perpendicular to the surface of substrate 21 in the central region. The direction of the current flow through the two spirals is such that these magnetic field components add.
The traces can be patterned on a wide variety of substrates. Substrates that are used in conventional printed circuit boards or flexible carriers are particularly attractive, as there is a well-developed technology for fabricating multiple layers of metal traces with selective connections between the traces on various layers. Printed circuit boards or circuit carriers are known to the art, and hence, will not be discussed in detail here. For the purposes of the present discussion it is sufficient to note that printed circuit boards can be fabricated by depositing thin metal layers, or attaching metal layers, on a somewhat flexible organic/inorganic substrate formed of fiberglass impregnated with epoxy resin and then converting the layers into a plurality of individual conductors by conventional photolithographic techniques.
Embodiments based on flex circuit technology are also attractive, as the substrates are inexpensive and can be provided with a thin substrate layer. The substrates are made of an organic material such as polyimide. Films and laminates of this type are available commercially from Dupont and utilize substrates called Kapton™ made from polyimide and, in some cases, a plurality of layers are provided with an adhesive. Embodiments in which other layers are provided by sputtering, or lamination are also available. In one embodiment, a Pyralux AP laminate from Dupont that has a 2 mils thick Kapton™ layer and copper layers on the top and bottom surfaces are utilized. In contrast to conventional printed circuit boards, flex carriers are flexible and can be bent to conform to various patterns.
Substrates made of other plastics or polymers can also be utilized depending on the particular application. In addition, inorganic substrates such as glass or ceramics could be utilized. The particular choice of substrate will, in general, depend on cost and the particular application. For example, glass and ceramic substrates are well suited for applications involving high voltages.
To simplify the following discussion, a component coil will be defined to be a substrate having a substantially planar insulating layer of material having top and bottom surfaces. The top surface includes a first trace having an outer end and an inner end and a first spiral conductor connected between said outer and inner ends of the first trace. As noted above, the spiral conductor includes a continuous and gradually widening linear conductor that forms a curve around a central region. The bottom surface includes a second trace having an outer end and an inner end and a second spiral conductor connected between said outer and inner ends of the second trace. A conductor connects the inner ends of the first and second traces. The central regions of the first and second spiral conductors overlie one another. The first and second spiral conductors are oriented such that a current traveling from the outer end of the first trace to the inner end of the first trace generates a magnetic field having a first component perpendicular to the top surface in the central region of that trace, and a current passing from the inner end of the second trace to the outer end of the second trace generates a magnetic field having a second component perpendicular to the top surface in the central region of the second trace, the first component having a direction that is the same as that of the second component. The outer ends of the first and second traces are accessed by power pads or wire bond pads that are part of the component coil.
Two or more of the component coils can be combined to provide a coil having additional windings. The component coils are combined by bonding the coils to one another and connecting the leads from the various component coils in the desired manner. Refer now to
The insulating layers will, in general, depend on the substrate used to construct the component coil. For example, in the case of a flexible carrier made from Kapton, the insulating layers can be provided by bonding a thin Kapton layer to the top and bottom surfaces using an insulating adhesive. If substrate 21 were constructed from glass or a ceramic, the insulating layers could be constructed by depositing a glass or ceramic layer over each surface of the substrate or Kapton could be used.
As noted above, two or more component coils can be connected together to provide a component coil having additional windings. Refer now to
While compound coils having traces connected in parallel have lower resistance, the need to drill and fill the vertical interconnects can pose problems, as the filling becomes more difficult as the hole aspect ratio (depth/diameter) increases. Hence, in some applications, it may be advantageous to use component coils that are connected in series.
Refer now to
Refer now to
The component coils can be combined to provide a transformer that has a primary and secondary winding. Refer now to
Embodiments in which the primary and/or secondary windings are constructed from a plurality of component coils can also be constructed. In this case, component coil 71 and/or component coil 72 shown in
In the above-described transformer embodiments, the component coils that made up the primary winding of the transformer were separated from those that made up the secondary winding of the transformer. However, embodiments in which the component coils that make up the primary and secondary windings are intermingled could also be constructed. Refer now to
The embodiments described above are analogous to air or dielectric core transformers. However, embodiments that incorporate magnetically-active materials such as ferrite, and in particular soft ferrite, can also be constructed. Refer now to
Refer now to
It should be noted that in embodiments in which space is a limiting factor, ferrite region 107 and the flux return layers 108 and 109 could be omitted. While the efficiency of energy transfer between the primary and secondary windings will be less efficient, such embodiments would still be better than embodiments that just utilize a non-ferrite core.
Transformers according to the present invention could be utilized to construct a galvanic isolator in which the components on one side of the isolation barrier are powered by a power source on the other side of the isolation barrier. Refer now to
Power section 150 includes a power supply 151 that powers the circuitry on both sides of the isolation gap. An inverter 152 generates an AC power signal from the DC power provided by power supply 151. The AC power signal is transferred to the receiver side of the isolation gap by a power transformer 153 according to the present invention. The secondary winding of power transformer 153 is rectified by converter 154 to provide a power supply 155 that is used to power receiver 163. It should be noted that the DC potentials provided by power supplies 151 and 155 could be the same or different, depending on the particular galvanic isolator design. Power transformer 153 can provide a voltage step up or step down to facilitate the generation of the different output voltages. It should also be noted that embodiments in which power is derived from a train of pulses applied to power transformer 153 from a source that is external to the galvanic isolator could also be constructed.
It should be noted that CMOS circuitry is not well adapted for rectifying AC power signals at high frequencies. Hence, converter 154 is preferably a separate component that is fabricated in a different integrated circuit system. However, if inverter 152 and transformer 153 are designed to operate at a frequency compatible with CMOS devices, the need for a separate component can be avoided. As pointed out above, the transformers of the present invention can be constructed using conventional circuit carriers or printed circuit boards. Hence, in one embodiment of the present invention, converter 154 is a separate circuit module that is located on the same circuit carrier as power transformer 153. Alternatively, the components of power section 150 and data transfer section 160 can be packaged in respective integrated circuit packages or together in a single larger integrated circuit package.
While galvanic isolator 140 utilizes a transformer for providing the data isolation gap, other forms of isolator could be utilized in combination with power section 150. The data isolation gap can be provided by a split circuit element in which one half of the element is on the transmitter side of the gap, and the other half is on the receiver side of the gap. For example, isolators based on optical links in which the transmitter generates a light signal that is received by a photodetector are known to the art.
A transformer according to the present invention can be constructed by stacking and bonding sheets of component coils. Refer now to
The above-described embodiments of the present invention could be modified to include traces and mounting pads for additional circuit elements. The transformers of the present invention already include structures analogous to conventional printed circuit board layers. Hence, providing attachment points for other circuit components is relatively inexpensive. As noted above, an attachment point for a power converter that rectifies the output of the secondary winding of the transformer is particularly useful. In addition, attachment pads for mounting other circuit components such as the receiver and transmitter die discussed above are also useful.
Refer now to
Data for transmission across the isolation gap provided by transformer 328 is input on bond pads 327 and 328 to a transmitter 322. Transmitter 322 is connected to the primary winding of transformer 328 by traces 321 and 325 in a manner analogous to that described above with respect to device 302. The secondary winding of transformer 328 is connected to receiver 323. The data from receiver 323 is coupled to a device external to galvanic isolator 300 via bond pads 327 and 326.
It should be noted that both transformer 318 and transformer 328 can be fabricated from the same stack of component coils 301. This further reduces the cost of galvanic isolator 300.
The above-described embodiments of the present invention utilize component coils for both the primary and secondary windings. However, embodiments in which one of the primary or secondary windings utilizes a coil or coils having only one spiral trace could also be constructed. In such embodiments the connection to the inner end of the spiral coil can be made either by a trace on another surface of the substrate or by a wire bond that is connected to the inner end of the spiral coil. Coils of this construction are discussed in detail in co-pending U.S. patent application Ser. No. 11/512,034 which is hereby incorporated by reference.
Refer again to
It should be noted that insulating layers 407 and 408 can be separately fabricated with the patterned ferrite layer thereon. Hence, the ferrite coupling feature can utilize the same basic component coil design and parts as non-ferrite component coils.
The above-described embodiments of the present invention utilize prefabricated component coils. However, embodiments in which the component coils are fabricated from individual coils during the fabrication of a transformer can also be constructed. Refer now to
Next, layers of polyamide resin are placed over the coils as shown at 455 and 456 in
Next, insulating overlays that have predrilled holes to provide openings overlying pads 463, 465, 464, and 467 are bonded to each of the exposed surfaces as shown at 491 and 492 in
As noted above, transformers according to the present invention are useful in constructing galvanic isolators that include two transformers, one for powering one of the receiver or transmitter and one for transmitting data. In some embodiments, the individual isolators may require shielding such that the magnetic field from one transformer is not coupled to the second transformer. For example, the power transformer, which generates a more intense magnetic field than the data transformer, could interfere with the data transmission if the alternating magnetic field generated in the power transformer is coupled to the data transformer. Such interference can be significantly reduced by providing a magnetic shielding layer on the top and bottom surfaces of the transformer.
In embodiments having a flux return loop such as the embodiments shown in
Shielding can also be provided by providing a separate layer of a magnetic shielding material such mumetal on the outer surface of each transformer. Refer now to
The galvanic isolators described above that utilize a transformer according to the present invention to provide power for one or more components in the isolator have utilized a single receiver and transmitter for the data path. However, galvanic isolators that include multiple data paths can also be constructed. Refer now to
Galvanic converter 600 includes two data transmission sections shown at 628 and 638. Data transmission section 628 includes a transmitter 622 and a receiver 623. Data transmission section 638 includes a transmitter 643 and a receiver 632. Receiver 623 and transmitter 643 are powered from the power converter in device 603.
Various modifications to the present invention will become apparent to those skilled in the art from the foregoing description and accompanying drawings. Accordingly, the present invention is to be limited solely by the scope of the following claims.
This is a continuation of application Ser. No. 12/392,978, filed on Feb. 25, 2009, which is a divisional application of application Ser. No. 11/747,092, filed on May 10, 2007, which are both hereby incorporated by reference for all that is disclosed therein.
Number | Date | Country | |
---|---|---|---|
Parent | 12392978 | Feb 2009 | US |
Child | 12709274 | US | |
Parent | 11747092 | May 2007 | US |
Child | 12392978 | US |