Decoupling capacitor

Figure 2-7 A Small Ceramic Decoupling Capacitor Needs to be Mounted Very Close to the Switcher 1C...

However, as you will see a little later, it is not a bad idea to always include this 0.1 pF input decoupling capacitor. The reasons may be different on different occasions, and for different types of switchers, but this component is generally always nice to have. 

In Figure 2-10, we Anally break up the input capacitance into a high-frequency capacitor and a relaAvely low-frequency bulk capacitor. The current distribuAons are shown, as well as how they all add up eventually. The mystery is clear now, and in the process we also understand how the decoupling capacitors are supposed to behave. Now we can also start to understand how this delicate balance can be easily shattered by lack of proper decoupling ... 

In DC-DC converters, we can somewhat slow the turn-on of Fets if we insert a small resistor (10 to 20Q. typically) in series with the decoupling capacitor of their respective driver stages. For example, a small resistor can be placed in series with the bootstrap capacitor of the third-generation switcher family I used to cover. That helped with almost 10 to 20% of customers, but somehow this trick didn t find its way into the applications information section of their datasheet. If the Fets are external, we can try a small resistor in series with the Gate, but this affects both the turn-on and turnoff (with such low threshold voltages, a diode in parallel to the resistor will not do anything). 

The ESR of the decoupling capacitors has to be minimized, as does the ESL. As can be seen from Figure 4-14, everything becomes important. In effect, this is the final search for the ideal capacitor Though there is no ideal capacitor, they certainly are getting better and better, as the next sections reveal. 

The gains for the popular lOOnF decoupling capacitor in a package of about 80mil x 50mil (0805 or similar) are summarized as follows ... 

For all topologies, that is in fact a key requirement—that the control IC be powered off a clean rail. It is just that in a Buck, the input decoupling capacitor for the power stage and the decoupling capacitor for the control are often the same component. Though in some cases, it may be necessary, even for a Buck controller IC, to add a small RC going to its supply pin (typically a 10Q resistor and an additional 0.1 pF ceramic capacitor). 

That may not sound too much, but has been known to cause problems in switcher ICs, especially those using Mosfets, for which an input ceramic decoupling capacitor for the IC becomes almost mandatory. Therefore, it is strongly advised that this capacitor be placed extremely close to where the pins of the IC actually contact the board, and further, there should be no intervening vias between this capacitor and the solder pads of the pins. 

Recommendation 5 (Figure 6-12) The bulk capacitor has been placed close to where the input supply leads came in, but the ceramic decoupling capacitor must be very close to the IC as indicated here. 

Figure 6-12 Input Decoupling Capacitor Placed Very Close to the IC Between VIN and GND Pins...

Figure 7-2 Box Mounting Constraints Can Force Non-optimum Primary-side Routing and Cause Excessive EMI—Possible Solution May Be to Mount a Decoupling Capacitor Close to the Fet...

We did it somehow, almost strangulating ourselves in the process. Now when I look back at this incident, I wonder why we didn t place a ceramic decoupling capacitor close to the switch, as shown in the lower half of the figure. The bulk capacitor could have successfully managed to provide the low-frequency current components, whereas the high-frequency capacitor could have really decreased the effective loop area in which the high-frequency components were circulating. 

So, if you can access the gates of the Fets, try putting in small resistors in series with them. If it is a switcher IC (with no access to the gates), try inserting a small resistor in series with the decoupling capacitor of the driver supply (usually a 0.1 pF capacitor attached to the Vdd pin and/or the bootstrap pin). Better still, pick an IC with less aggressive drive to start with. Because otherwise it will commit suicide sooner or later. 

Figure J shows an example of the top surface features of an MCP designed for electrooptical-signal-processing applications (33). The MCP has 18 chip attach pads surrounded by dumbbell-shaped pads for wire bonding and repair. The top surface also contains off-package I/Os along two sides, wide power distribution lines, and sites for decoupling capacitors. In this design, the package size of 2.25 by 2.25 in. (5.7 by 5.7 cm) was determined by the top-layer features rather than by the maximum interconnection density.

In the multichip package, a variety of pretested chips (e.g., bipolar, MOS, and GaAs) and discrete components (decoupling capacitors and termination resistors) may be mounted on the high-density interconnection substrate. This approach is sometimes termed hybrid-wafer-scale integra-... 

Applications that have received attention, and the material properties that enable them, are shown in Figure 27.1. These applications are reviewed in detail in Waser and Ramesh. Decoupling capacitors and filters on semiconductor chips, packages, and polymer substrates (e.g., embedded passives ) utilize planar or low aspect ratio oxide films. These films, with thicknesses of 0.1 to 1 J,m, are readily prepared by CSD. Because capacitance density is a key consideration, high-permittivity materials are of interest. These needs may be met by morpho-tropic phase boundary PZT materials, BST, and BTZ (BaTi03-BaZr03) solid solutions. Phase shifters (for phase array antennas) and tunable resonator and filter applications are also enabled by these materials because their effective permittivity exhibits a dependence on the direct current (DC) bias voltage, an effect called tunability. 

---