SLVSE83B December 2020 – September 2023 TPS6593-Q1
PRODUCTION DATA
The buck converters have seven potential NVM configurations which can impact the output capacitor selection. Refer the part number specific user's guide to identify which configuration applies to each buck regulator. The actual minimal capacitance requirements to achieve a specific accuracy or ripple target varies depending on the input voltage, output voltage, and load transient characteristics; some guidance, however, is provided below. The local output capacitors must be placed as close to the inductor as possible to minimize electromagnetic emissions. Every buck output requires a local output capacitor to form the capacitive part of the LC output filter. It is recommended to place all large capacitors near the inductor. See Section 9.4 for more information about component placement. Use ceramic local output capacitors, X7R or X7T types; do not use Y5V or F. DC bias voltage characteristics of ceramic capacitors must be considered. The output filter capacitor smooths out current flow from the inductor to the load, helps maintain a steady output voltage during transient load changes and reduces output voltage ripple. These capacitors must be selected with sufficient capacitance and sufficiently low ESR and ESL to perform these functions. Minimum effective output capacitance (including the DC voltage roll-off, tolerances, aging and temperature effects) is defined in Electrical Characteristics table for different buck configurations. The output voltage ripple is caused by the charging and discharging of the output capacitor and also due to its RESR. The RESR is frequency dependent (as well as temperature dependent); make sure the value used for selection process is at the switching frequency of the part.
To achieve better ripple and transient performance, additional high pass filter caps are recommended to compensate for the parasitic impedance due to board routing and provide faster transient response to a load step. These caps are placed close to the point of load and are also the input capacitors of the load. These capacitors are referred to as POL caps later in this document. POL capacitor usage varies based on the application and generally follows the SoC or FPGA input capacitor requirements. Low ESL 3-terminal caps are recommended, as their high performance can help reduce the total number of capacitors required which simplifies board layout design and saves board area. They also help to reduce the total cost of the solution.
Note that the output capacitor may be the limiting factor in the output voltage ramp and the maximum total output capacitance listed in electrical characteristics must not be exceeded. At shutdown the output capacitors are discharged to 0.15-V level using forced-PWM operation. This discharge of the output capacitors can cause an increase of the input voltage if the load current is small and the output capacitor is large. Below 0.15-V level the output capacitor is discharged by the internal discharge resistor and with large capacitor more time is required to settle VOUT down as a consequence of the increased time constant.
Figure 9-3 is an example power distribution network (PDN) of local and POL caps at the output of a buck for optimal ripple and transient performance. Table 9-6 lists the local and POL capacitors used to validate the buck transient and ripple performance specified in the parametric table for each of the seven configurations. Table 9-7 lists the actual capacitor part numbers used for the different use case tests, neglecting capacitors below 10-µF. It is recommended to simulate and validate that the capacitor network chosen for a particular design meets the desired requirements as these are provided as guidelines.
Configuration | COUT | L | CL / phase | RPCB per phase1 | LPCB per phase2 | CPOL1 (total) | CPOL2 (total) |
---|---|---|---|---|---|---|---|
4.4 MHz VOUT Less than 1.9 V, Low Load Step, Single Phase with low COUT |
Low |
470 nH |
22 uF x 1 |
8 mΩ | 2.5 nH |
1 uF x 1 |
|
4.4 MHz VOUT Less than 1.9 V, Multiphase | Low | 220 nH | 47 µF × 2 | 8 mΩ | 2.5 nH | 10 µF × 4 | |
High | 220 nH | 47 µF × 4 | 8 mΩ | 2.5 nH | 10 µF × 2 | ||
4.4 MHz VOUT Less than 1.9 V, Single Phase with high COUT | Low | 220 nH | 47 µF × 1 | 8 mΩ | 2.5 nH | 10 µF × 4 | |
High | 220 nH | 47 µF × 4 | 8 mΩ | 2.5 nH | 10 µF × 2 | ||
4.4 MHz VOUT Less than 1.9 V, Single Phase with low COUT | Low | 220 nH | 22 µF × 1 | 8 mΩ | 2.5 nH | 10 µF × 2 | |
High | 220 nH | 47 µF × 1 | 8 mΩ | 2.5 nH | 10 µF × 4 | ||
4.4 MHz VOUT Greater than 1.7 V, Single Phase Only (VIN Greater than 4.5 V) | Low | 470 nH | 47 µF × 1 | 27 mΩ | 6 nH | 10 µF × 4 | |
High | 470 nH | 47 µF × 2 | 27 mΩ | 6 nH | 10 µF × 2 | ||
2.2 MHz Full VOUT Range and VIN Greater than 4.5 V, Single Phase Only | Low | 1000 nH | 47 µF × 3 | 8 mΩ | 2.5 nH | 10 µF × 4 | |
High | 1000 nH | 47 µF × 3 | 8 mΩ | 2.5 nH | 10 µF × 4 | 680 µF × 1 | |
2.2 MHz VOUT Less than 1.9 V Multiphase or Single Phase | Low | 470 nH | 47 µF × 3 | 4.1 mΩ | 1.3 nH | 10 µF × 4 | |
High | 470 nH | 47 µF × 3 | 4.1 mΩ | 1.3 nH | 10 µF × 4 | 680 µF × 1 | |
2.2 MHz Full VOUT and Full VIN Range, Single Phase Only | Low | 1000 nH | 47 µF × 3 | 4.1 mΩ | 1.3 nH | 10 µF × 2 | |
High | 1000 nH | 100 µF × 4 | 4.1 mΩ | 1.3 nH | 10 µF × 2 | ||
DDR VTT Termination, 2.2 MHz Single Phase Only | - | 470 nH | 22 µF × 1 | 27 mΩ | 6 nH | 10 µF × 1 + 22 µF x 1 |
Power input and output wiring parasitic resistance and inductance must be minimized.
MANUFACTURER | PART NUMBER | VALUE | EIA Size Code | SIZE (mm) | Used for Validation |
---|---|---|---|---|---|
Murata | NFM15HC105D0G(1) | 1 µF, 4 V, X7S | 0402 | 1.0 × 0.5 | Yes |
TDK | YFF18AC0J105M(1) | 1 µF, 6.3 V | 0603 | 1.6 × 0.8 | - |
Murata | NFM18HC106D0G(1) | 10 µF, 4 V, X7S | 0603 | 1.6 × 0.8 | Yes |
TDK | YFF18AC0G475M(1) | 4.7 µF, 6.3 V | 0603 | 1.6 × 0.8 | - |
Murata | GCM31CR71A226KE02 | 22 µF, 10 V, X7R | 1206 | 3.2 × 1.6 | Yes |
Murata | GCM21BD7CGA5L1X7R0J226MT0J226M | 22 µF, 6.3 V, X7T | 0805 | 2.0 × 1.25 × 1.25 | - |
TDK | CGA5L1X7R0J226MT | 22 µF, 6.3 V, X7R | 1206 | 3.2 × 1.6 | - |
TDK | CGA4J1X7T0J226MT | 22 µF, 6.3 V, X7T | 0805 | 2.0 × 1.25 × 1.25 | - |
Murata | GCM32ER70J476ME19 | 47 µF, 6.3 V, X7R | 1210 | 3.2 × 2.5 | Yes |
Murata | GCM31CD70G476M | 47 µF, 4 V, X7T | 1206 | 3.2 × 1.6 | - |
TDK | CGA6P1X7S1A476MT | 47 µF, 10 V, X7S | 1210 | 3.2 × 2.5 | - |
TDK | CGA5L1X7T0G476MT | 47 µF, 4 V, X7T | 1206 | 3.2 × 1.6 | - |
Murata | GCM32ED70G107MEC4 | 100 µF, 4 V, X7S | 1210 | 3.2 × 2.5 | Yes |
TDK | CGA6P1X7T0G107MT | 100 µF, 4 V, X7T | 1210 | 3.2 × 2.5 | - |
Kemet | T510X687K006ATA023(2) | 680 µF, 6.3 V | 2917 | 7.4 × 5.0 | Yes |
Murata |
GCM188D70E226ME36D |
22uF, 2.5 V, X7T |
0603 |
1.6 × 0.8 |
Yes |