I often receive questions about why we place a physically large tantalum capacitor on the input of most low-power, step-down converter TPS62xxx evaluation modules (EVMs). Tantalum capacitors are generally very large and tall. Plus, in many applications, they simply cost too much or are not reliable enough to use in a robust system. So why do we have them on our EVMs?
An EVM is designed to be a simple test vehicle for an integrated circuit (IC) and its supporting components. Customers should be able to simply wire up a power supply, load, and set a few jumpers to begin their evaluation. This is quite different than a real system, which generally requires careful component selection and settings (and many reviews of each) to ensure sufficient performance.
In order to support a simple evaluation, the tantalum input capacitor ensures a steady, low-impedance supply voltage for the IC. This is how a typical system is implemented in its final configuration. The battery resides next to the printed circuit board (PCB) and is soldered with very short wires before its voltage enters the PCB traces. For a system powered by an adapter, an input filter is often installed between the adapter voltage and the system voltage. This filter usually has a large amount of capacitance in order to comply with conducted emissions standards. Both the battery-powered system and adapter-plus-filter-powered system are different from the typical lab test setup, which just has long wires connecting a standard lab power supply to the EVM. But such a connection adds significant impedance to the input supply.
This impedance, which is not typically present in the final system, can cause difficulties when evaluating the EVM. For instance, a load transient at the output will cause a sudden increase in input current. If the impedance of the connection between the input supply and EVM is too high, a temporary drop of the input voltage occurs and the resulting load transient performs worse than it would with a steady input voltage. Thus, that steady input voltage comes from the large tantalum capacitor on our EVMs.
A tantalum capacitor also has some equivalent series resistance (ESR). This greatly helps in snubbing the inductor/capacitor (LC) resonant circuit formed by long wires and input capacitance in a lab environment. The ESR – combined with the additional capacitance of the tantalum capacitor – greatly reduces voltage overshoots caused by “hot-plugging” the input supply to the EVM. Hot-plugging is when the input power supply is enabled and its wires are connected to the EVM. This places a “hot” (live) 12V on the EVM. The voltage then rings through the inductance of the cable and causes an overshoot on the EVM. An example of this is shown in Figure 1, taken from the TPS62125 data sheet. The first waveform has the potential to damage the device, while the second is safe.
Figure 1: Hot-plugging creates an input voltage transient, which may damage the IC
Since the tantalum capacitor is there merely to enable easy evaluation on the EVM, it is not typically required in the final implementation. If there is significant impedance in the input supply in the final application, some additional input capacitance may be required. The exact value of the required capacitance is application-dependent and should be uncovered with lab testing. In most applications, no additional capacitance is required due to a low input-impedance.
And that’s why we put tantalum capacitors on our EVMs. Do you have other questions about why we do what we do on EVMs? Ask us here.
