The planar integrated LC structure consists of alternating layers of conductors, dielectrics, insulation and ferrite materials that produce an integrated structure with similar terminal characteristics as the lumped components. The exploded view of an integrated LC structure was shown in Fig. 1-4(a).The integrated LC winding consists of a dielectric substrate with conductor windings directly deposited on both sides, thus resulting in a structure having both sufficient inductance and capacitance. This realizes the equivalent integrated capacitance as well as the inductance. By appropriately terminating the four terminals A, B, C and D of the integrated LC winding, the same structure could be configured as equivalent LC series resonator, parallel resonator or lowpass filter. To integrate the EMI filter, the LC lowpass filter configuration is used, where AD is the input port and CD is the output.
The existing integrated LC technologies and design methodologies were mostly developed for high-frequency power passive components integration in order to achieve high efficiency and high power density. Since functions and requirements are different for passive components in EMI filters, special technologies need to be developed for EMI filter integration.
3.2. Functions and requirements for HF power passive components and EMI
According to the fundamental functionality performed, a typical power electronics converter can be divided into several function blocks, as shown in Fig. 3-1 . The switching function is used to control the flow of electromagnetic energy through the power processor (converter), the conduction function is used to direct the flow appropriately, while the electromagnetic energy storage function provides energy continuity at the input and output of the processor. The information function takes care of the necessary sensing and relates the spatial and temporal action of the first three
functions. Finally, the heat exchange function maintains the thermal integrity of the system.
The passive components, including transformers, inductors and capacitors, can be categorized into the electromagnetic energy storage function block. Within this block, the HF power passive components should store and propagate the electromagnetic energy at the switching frequency. Meanwhile, the EMI filters, which also consist of passive components, should attenuate the electromagnetic energy at the switching frequency and its harmonics. Therefore, the requirements and constraints for EMI filters are different from those of other HF passive components.
For HF power passive components, the major constraint is thermal, which eventually limits the component size. Although integration helps to reduce the component size and increase the power density, the loss density also increases. To alleviate the thermal problem, the integrated module normally needs to be designed to minimize the losses, especially for switching frequency and above. Also, since the important frequency range of these modules is close to their self-resonant frequency, the high-frequency characteristics are normally not of great concern.
However, for EMI filters, the requirements are quite different. Firstly, since the function of the EMI filter is to attenuate the noise at switching frequency and the harmonics; the fundamental self-resonant frequency of the integrated EMI filter, which is the corner frequency of the lowpass filter, is normally much lower than the switching frequency. The high-frequency performance now becomes a major concern for the component and filter design. Secondly, in terms of losses, high-frequency losses are desired to dampen the high-frequency noise. Since the power current passing through the
filter is at line frequency, increasing high-frequency losses will not worsen the thermal problem, as long as the losses at line frequency are kept small. From the above analysis, it can be concluded that the major technology requirement for the integrated EMI filter is to improve high frequency performance by reducing EPC and ESL of the integrated inductors and capacitors, and by increasing-high frequency losses.
3.3. Material, electromagneti,c structural, and processing limitations
To achieve improved characteristics of EMI filters via an integration approach, the
material, electromagnetic, structural, and processing limitations have to be considered.
3.3.1. Material limitations
The materials used in EMI filters include conductor materials (Cu), high-permeability magnetic materials (ferrite) for CM chokes, low-permeability material for DM chokes, high-permittivity dielectric material for DM capacitors, medium-permittivity dielectric material for CM capacitors, and low-permittivity dielectric material for insulations. To obtain good performance over a wide frequency, temperature, and excitation field range, good linearity of the magnetic and dielectric materials is desirable. However, the highpermeability or high-permittivity material normally has very limited frequency, temperature, and excitation field characteristics. These problems have to be considered when designing EMI filters. The characteristics of commonly used conductor, magnetic and dielectric materials are listed in Table 3-1.
3.3.2. Electromagnetic limitations
For any structure with finite volume, there exists stray electromagnetic field. The energy stored in the stray electric field is represented by structural capacitance, while the energy stored in the stray magnetic field is represented by structural inductance. Selfresonance of a component is caused by a temporal and spatial energy exchange between the electric and magnetic fields. Since integration cannot eliminate the stray electromagnetic field, it can never eliminate the parasitics. This is a physical law with which electrical engineers must contend. However, with an appropriately-designed integrated structure and layout, the energy distribution can be changed so that the energy exchange between electric and magnetic fields can be manipulated. If a zero net-energy exchange can be achieved, a perfect component can be constructed. This will be a research subject we are going to study.
3.3.3. Structural limitations
The structural integrity of discrete components is achieved by packaging every component separately. When integrating the passive components into a single module, they are structurally and geometrically coupled. Each part of the integrated module needs to mechanically support the other parts. The integrated inductor and the integrated capacitor share the same conductor and cannot be separated. The interconnection of different parts within the module also needs to be carefully designed. To obtain a mechanically and electrically stable module, structural limitations have to be considered.
3.3.4. Processing limitations
To construct the integrated EMI filters, standard semiconductor and printed circuit board processing technologies will be used to reduce the labor necessary during the processing steps. The major processing technologies are: direct metallization (sputtering), electro-plating, photolithography, chemical or plasma etching, reflow soldering, and laser cutting. The limits of the current processing technologies are listed in Table 3-2.
Because of existing material, electromagnetic, structural and processing concerns, the characteristics of discrete and integrated EMI filters are limited. To overcome these limitations, special technologies suitable for EMI filter integration have to be developed.
3.4. Implementation of integrated
EMI filters As mentioned in the previous section, the fundamental element of an EMI filter is a lowpass filter, for both CM and DM excitations. To realize the integrated lowpass filter, the integrated L-C structure is analyzed and its low-pass filter configuration and equivalent circuit are derived.
3.4.1. Lowpass filter configuration of integrated
L-C structure The integrated L-C structure is a distributed parameter structure, with conductive and displacement current existing at the same time. Neglecting losses, the simplified equivalent circuit is shown in Fig. 3-2(a). For the lowpass configuration shown in Fig. 3-2(b), where AD is the input port and CD is the output port, the conduction and displacement current distribution is illustrated in Fig. 3-2(c). If an infinitesimal segment of the integrated L-C winding is considered, as shown in Fig. 3-3, the voltage drop along the top line (V12) and the bottom line (V1’2’) are given by:
where L is the self-inductance per unit length, M is the mutual inductance per unit length. Since the top and bottom conductors are placed very close together, M ≈ L . According to Kirchhoff’s current law (KCL), in i (x) + i (x) = I 1 2 . Equation (3-1) and (3-2) can be simplified to
Knowing this equation, the voltage distribution along the length of the integrated L-C structure can be determined, as shown in Fig. 3-4, where
Since along the winding length, the voltage on both top and bottom conductors vary linearly with the same slope, the voltage difference between the top and bottom lines (V11’, V22’…) is a constant, which is defined as:
the voltage difference between the top and bottom lines
The total displacement current is:
And the output current is equal to: Io=Iin – Ic
When comparing this to the terminal characteristics of an L-C low pass filter, shown
in Fig. 3-5, where
Iin = Ia
It is evident that when L- X = Lf ⋅ and C- X = Cf ⋅ , the distributed network has the same terminal characteristics as the lumped L-C lowpass filter, under the first order approximation. This low pass filter configuration of the integrated L-C structure will be the fundamental function block for EMI filter integration.
3.4.2. Implementation of integrated CM filters
Under CM excitation, the EMI filter can be simplified to two low pass filters in parallel. Hence, the integrated CM filter can be realized by two integrated L-C windings, as shown in Fig. 3-6. In Fig. 3-6, the two integrated L-C windings are both configured as lowpass filters and they are closely magnetically coupled. Under the first order approximation, the equivalent circuit corresponding to Fig. 3-6 is shown in Fig. 3-7
3.4.3. Implementation of integrated
DM filters The equivalent circuit of DM filters is a π-type lowpass filter, which has a small filter inductance (10 µH – 20 µH) and two large filter capacitances (0.1 µF – 1 µF). Similar to the discrete EMI filters, the DM inductance can also be realized by utilizing the leakage inductance of the integrated CM choke. The planar CM choke provides more flexibility for controlling the leakage inductance, which is achieved by inserting additional magnetic material between windings, as illustrated in Fig. 3-8. Without changing the number of
turns of the CM inductor, the leakage inductance can be varied by tuning the permeability and effective area of the inserted magnetic material. This gives an opportunity to decouple the design of the CM and DM inductors. The DM capacitor can be implemented by another integrated LC winding connected as capacitors. It can be a simple, one-turn or a partial-turn winding, as shown in Fig. 3-9.
3.4.4. Integrated EMI filter implementation
The schematic to illustrate the integrated EMI filter composition is shown in Fig. 3-10. The exploded view of the physical structure is shown in Fig. 3-11.