The present invention relates to a component arrangement with a planar transformer.
One component arrangement is described, by way of example, in DE 102 32 642 A1.
In this component arrangement, a dielectric layer 120 is arranged on a semiconductor body 110 and electrically isolates a primary winding 140 and a secondary winding 130 of a planar transformer from one another. The secondary winding 130 is connected, for example, to integrated circuit components (which are not illustrated in any more detail) in the semiconductor body. The primary winding may be connected to other circuit components in the same semiconductor body 110 or in another semiconductor body (not illustrated). The circuit components to which the primary winding 150 is connected form, in particular, a transmission circuit, and the components to which the secondary winding is connected form, in particular, a receiving circuit for a data transmission device, in which the transformer is used as an inductive coupling element between the transmitter and receiver, and at the same time as a potential barrier between the transmitter and receiver.
The primary winding 140 and the secondary winding 130 are each arranged as a conductor loop with two or more turns on in each case one (metallization) level in the dielectric layer 120, and thus form a planar transformer without a transformer core, which is referred to in the following text as a coreless transformer.
In the equivalent circuit shown in
Parasitic effects also result in capacitive coupling between the primary winding 140 and the secondary winding 130. C134/2 in
Coreless transformers of the type explained above are used, for example, in half-bridge circuits for the transmission of a drive signal from a control circuit to a high-side switch in the half-bridge circuit, in order to decouple the potentials in the drive circuit and in the high-side switch. In circuit arrangements such as these, electromagnetic interference signals occur during switching processes of the high-side and low-side switches which form the half-bridge circuit and are normally in the form of power transistors, and these interference signals can induce interference voltages in the windings of the transformer. These interference voltages are produced by displacement currents in the parasitic capacitances between the primary winding and the secondary winding and may, in some circumstances, reach the level of useful signals to be transmitted.
In conventional iron-core transformers, which have been known for a long time, the effect of parasitic capacitances is reduced by the use of a shielding layer between the primary winding and the secondary winding of the transformer.
In so-called pulse transformers, which are used for signal transmission, the primary winding and the secondary winding are arranged as far apart from one another as possible on a torroidal annular core, although this does not significantly reduce the parasitic capacitances, since, as before, there is still a large capacitance between the windings and the annular core.
Differential transmission methods are known for signal transmission using planar coreless transformers, and these allow detection of interference signals which are injected into the transmission path. Methods such as these are described, by way of example, in DE 102 29 860 A1. These transmission methods are, however, comparatively complex.
One aim of the present invention is to provide a component arrangement with a planar transformer which is robust against electromagnetic interference signals when used in a signal transmission path.
The component arrangement according to the invention has, according to a first aspect of the present invention:
In the component arrangement according to the invention, the splitting of the secondary winding into a first and a second winding section, with one of the two winding sections being arranged between the primary winding and the other of the two winding sections, leads to a reduction in the parasitic capacitance between the primary winding and the secondary winding, and makes the component arrangement according to the invention more robust against electromagnetic interference, in comparison to conventional component arrangements with planar transformers, when using the component arrangement in a signal transmission path.
One embodiment of the invention provides for the first and second winding sections of the secondary winding each to have more than one turn. A winding sense of the first winding section in this case preferably runs in the opposite direction to a winding sense of the second winding section.
A further aspect of the invention provides for the one winding section of the secondary winding, which is arranged between the other winding section and the primary winding, to have one and only one turn, one of whose ends is separated from its other end by a gap. The dimensions of this one winding section in a lateral direction in this case correspond at least approximately to the dimensions of the other winding section in the lateral direction, or to the dimensions of the primary winding in the lateral direction.
A further aspect of the invention relates to a component arrangement which has the following features:
In this component arrangement, the third winding forms a shield between the primary winding and the secondary winding, and thus ensures that the parasitic capacitance between the primary winding and the secondary winding is reduced, thus resulting in increased robustness of the component arrangement against electromagnetic interference radiation when used in a signal transmission path.
The present invention will be explained in more detail in the following text using exemplary embodiments and with reference to figures, in which:
Unless stated to the contrary, identical reference symbols in the figures denote identical components and their parts that have the same meaning.
With reference to
The secondary winding 30 of the component arrangement has two winding sections, specifically a first winding section 31 and a second winding section 32, which is arranged at a distance from the first winding section 31 in a vertical direction of the semiconductor body 10 and of the dielectric layer 20. The second winding section 32 is in this case arranged in the vertical direction between the first winding section 31 of the secondary winding 30 and the primary winding 40 in the dielectric layer 20. The dielectric layer 20 is composed, for example, of a semiconductor oxide, in particular silicon oxide. However, any desired further electrically isolating layers may, of course, also be used as the dielectric layer 20.
In the illustrated example, the first winding section 31 is located immediately adjacent to the semiconductor body 10, but with the individual turns being arranged so that they are isolated from the semiconductor body 10. An electrically conductive connection between the secondary coil 30 and the circuit components of the semiconductor body 10 is made—provided that this is desired—in a manner which is not illustrated in any more detail, via connections 34, 36 of the secondary winding.
The primary winding 40 and the two winding sections 31, 32 of the secondary winding each have two or more turns, which are arranged in the form of a spiral on one level, as is illustrated in
The primary winding 40 has a first and a second end 41, 42, which each form connections of this primary winding 40. In a corresponding manner, the first planar winding section 31 and the second planar winding section 32 of the secondary winding each have first ends 34, 36, which form first connections of these two winding sections 31, 32, and each have second ends 35, 37, which form second connections of the two winding sections 31, 32. The first connections 34, 36 of the first and second winding sections 31, 32 form connections of the secondary winding 30 at which a voltage which is induced in the secondary winding 30 by the primary winding 40 can be tapped off. The connections of the secondary winding 30 are in each case formed by the “outer” connections 34, 36, that is to say the connections 34, 36 which are located on the outside in the lateral direction on the spiral winding sections 31, 32. The “inner” connections 35, 37 of the winding sections 31, 32 are formed by an electrically conductive connection 33 which runs in places in the vertical direction between a level on which the first winding section 31 is formed and a level on which the second winding section 32 is formed.
These levels on which the first and second winding sections 31, 32 of the secondary winding 30 and of the primary winding 40 as well are formed are preferably so-called wiring levels in the dielectric layer 20. These wiring levels are produced, in a manner which has been known for a long time, by successively depositing two or more layer elements of the dielectric layer 20 one above the other, in which case cutouts can be produced in each of these layer elements by means of masking and etching processes which have been known for a long time, with these cutouts being filled with an electrically conductive material before the next layer element is deposited. The structures composed of electrically conductive material form, for example, wiring for components which are arranged in the semiconductor body 10, in which case the wiring on individual levels can be connected to one another by means of vertically running connections, so-called vias. The illustrated spiral windings and winding sections may be produced by spiral structuring of the individual mask layers, in which case the windings can be connected to the semiconductor body 10 through vias.
d shows the electrical equivalent circuit of the already explained component arrangement. In the equivalent circuit, the connections which correspond to the connections of the windings or winding sections 31, 32, 40 in
In the equivalent circuit, C40 denotes the capacitance of the primary winding, which acts between the connections 41, 42 of the primary winding. R40 denotes the resistance of the primary winding 40, (1-k)·L40 denotes the inductance value of any stray inductance which results from the inductance L40 of the primary winding, and k·L40 denotes the inductance value of that component of the inductance L40 of the primary winding which is involved in the magnetic coupling. The resistance R40, the stray inductance (1-k)·L40 and the coupling inductance k·L40 form a series circuit between the connections 41, 42, which is connected in parallel with the winding capacitance C40. In the equivalent circuit, C31 denotes the capacitance of the first winding section 31 of the secondary winding, R31 and L31 denote the resistance and the inductance of this first winding section 31, and they form a series circuit in parallel with the capacitance C31. In a corresponding manner, C32 denotes the capacitance of the second winding section 32 of the secondary winding 30, and R32 and L32 denote the resistance and the inductance of this second winding section, and they form a series circuit in parallel with the capacitance C32. The total input capacitance between the connections 34, 36 of the secondary winding 30 is denoted C3132, and this is considerably greater than the individual capacitances C31, C32 of the winding sections 31, 32 as a result of the short distance in the vertical direction between the winding sections 31, 32.
In
Any coupling capacitance between one of the connections 41, 42 of the primary winding 40 and the connection 36 of the secondary winding 30 is effectively negligible owing to the shielding effect of the second winding section 32, and is therefore not included in the equivalent circuit. That component of the coupling capacitance C4032/2 which acts between the connection 42 of the primary winding 40 and the connection 34 of the secondary winding has no effect on the signal transmission provided that the connection 34 of the secondary winding is connected to a reference ground potential, as is assumed in the equivalent circuit. Electromagnetic interference which is injected via the component of the coupling capacitance C4032/2 which acts between the connections 41 and 33 affects only the resistance component of the second winding section R32 and the inductive component L32.
a shows a cross-sectional side view of a further exemplary embodiment of the component arrangement according to the invention, in which the second winding section 32, which is arranged on a wiring level between the first winding section 31 of the secondary coil 30 and the primary winding 40, has one and only one turn.
With reference to
In the exemplary embodiment illustrated in
The dimensions of the second winding section 32, which has one turn, in the lateral direction are chosen such that they correspond at least approximately to the dimensions of the first winding section 31, which has two or more turns, such that the second winding section 32 shields the first winding section 31 of the secondary winding 30 from the primary winding 40.
In a further exemplary embodiment, which is illustrated in
A third planar winding 250, which has only one turn, is arranged between the primary winding 240 and the secondary winding 230, on the dielectric layer, for example on a further wiring level, with a first end 251 and a second end 252 of this turn being separated by a gap 253 which is filled by the material of the dielectric layer. This third winding 250 is operated with an open circuit, that is to say its ends 251, 252 are not connected. The third winding 250 is either at a floating potential or is connected to a reference ground potential, for example to the reference ground potential to which the semiconductor body 210 located underneath it is also connected. This is normally the reference ground potential to which the rear face of the semiconductor body 210, facing away from the dielectric layer 220, is also connected.
A plan view of the geometry of the primary winding 240 corresponds, for example, to the geometry of the primary winding 40 shown in
c shows the electrically equivalent circuit of the component arrangement which has already been explained with reference to
In this equivalent circuit, C240 and C230 denote the capacitances of the primary winding C240 and of the secondary winding 230. R240 and R230 denote the resistances of the primary winding 240 of the secondary winding 230. L240 and L230 denote the inductances of the primary winding 240 and of the secondary winding 230, with k·L240 and, respectively, k·L230, denoting the coupling inductances which result from these inductances, and (1-k)·L240 and, respectively, (1-k)·L230 denoting the respective stray inductances. The resistance R240, R230 as well as the stray and coupling inductances in each case form a series circuit, which is connected in parallel with the respective capacitance C240, C230 of the windings 240, 230. In
It is preferable for no further components to be provided under the windings 230, 240 in the semiconductor body 210. In this case, the semiconductor body 210 underneath the windings 230, 240 is composed entirely just of material of one conductance type, for example of p-conductive semiconductor material. The semiconductor body 210 then represents a conductive connection between the parasitic substrate capacitances Csub/2 and the rear face of the semiconductor body 210, which is normally at a reference ground potential. This reference ground potential is denoted by GND in
As can be seen from the equivalent circuit, the third winding 250 means that there is no capacitive coupling between the connections 241, 242 of the primary winding and the connections 234, 236 of the secondary winding. The equivalent circuit is based on the assumption that the third winding 250 is connected to a reference ground potential GND2, such that, in this component arrangement, only parasitic capacitances which are denoted by Cs/2 exist between the first and second connections 241, 242 of the primary winding 240 and this reference ground potential. This reference ground potential GND2 may correspond to the reference ground potential GND to which the parasitic substrate capacitances Csub/2 are also connected. However, these reference ground potentials GND, GND2 may also differ. It is thus possible to arrange a DC voltage source between these two reference ground potentials GND, GND2, or a capacitor whose capacitance is very large in comparison to the capacitances Csub/2.
Number | Date | Country | Kind |
---|---|---|---|
10 2004 036 139 | Jul 2004 | DE | national |
Number | Name | Date | Kind |
---|---|---|---|
4785345 | Rawls et al. | Nov 1988 | A |
4816784 | Rabjohn | Mar 1989 | A |
5572179 | Ito et al. | Nov 1996 | A |
5781071 | Kusunoki | Jul 1998 | A |
6097273 | Frye et al. | Aug 2000 | A |
6794977 | Christensen | Sep 2004 | B2 |
6927662 | Kahlmann et al. | Aug 2005 | B2 |
20040005009 | Taghizadeh-Kaschani | Jan 2004 | A1 |
20040056749 | Kahlmann et al. | Mar 2004 | A1 |
20040140528 | Kim et al. | Jul 2004 | A1 |
Number | Date | Country |
---|---|---|
102 29 860 | Jan 2004 | DE |
102 32 642 | Feb 2004 | DE |
11307723 | Nov 1999 | JP |
WO 9850956 | Nov 1998 | WO |
Number | Date | Country | |
---|---|---|---|
20060028313 A1 | Feb 2006 | US |