10.2.1.2.2.1 Surface Mount Inductors

Working inductance decreases as inductor current increases. If the drop in working inductance is severe enough, it may cause the boost converter to become unstable, or cause the TPA2013D1 to reach its current limit at a lower output power than expected. Inductor vendors specify currents at which inductor values decrease by a specific percentage. This can vary by 10% to 35%. Inductance is also affected by DC current and temperature.

10.2.1.2.2.2 TPA2013D1 Inductor Equations

Inductor current rating is determined by the requirements of the load. The inductance is determined by two factors: the minimum value required for stability and the maximum ripple current permitted in the application.

Use Equation 2 to determine the required current rating. Equation 2 shows the approximate relationship between the average inductor current, I_L, to the load current, load voltage, and input voltage (I_CC, V_CC, and V_DD, respectively). Insert I_CC, V_CC, and V_DD into Equation 2 to solve for I_L. The inductor must maintain at least 90% of its initial inductance value at this current.

Equation 2.

The minimum working inductance is 2.2 μH. A lower value may cause instability.

Ripple current, ΔI_L, is peak-to-peak variation in inductor current. Smaller ripple current reduces core losses in the inductor as well as the potential for EMI. Use Equation 3 to determine the value of the inductor, L. Equation 3 shows the relationship between inductance L, V_DD, V_CC, the switching frequency, f_boost, and ΔI_L. Insert the maximum acceptable ripple current into Equation 3 to solve for L.

Equation 3.

ΔI_L is inversely proportional to L. Minimize ΔI_L as much as is necessary for a specific application. Increase the inductance to reduce the ripple current.

Select an inductor with a small DC resistance, DCR. DCR reduces the output power due to the voltage drop across the inductor.

10.2.1.2.3.1 Surface Mount Capacitors

Temperature and applied DC voltage influence the actual capacitance of high-K materials.

Table 6 shows the relationship between the different types of high-K materials and their associated tolerances, temperature coefficients, and temperature ranges. Notice that a capacitor made with X5R material can lose up to 15% of its capacitance within its working temperature range.

High-K material is very sensitive to applied DC voltage. X5R capacitors can have losses ranging from 15 to 45% of their initial capacitance with only half of their DC-rated voltage applied. For example, if 5 Vdc is applied to a 10-V, 1-μF X5R capacitor, the measured capacitance at that point may show 0.85 μF, 0.55 μF, or somewhere in between. Y5V capacitors have losses that can reach or exceed 50% to 75% of their rated value.

In an application, the working capacitance of components made with high-K materials is generally much lower than nominal capacitance. A worst-case result with a typical X5R material might be –10% tolerance, –15% temperature effect, and –45% DC-voltage effect at 50% of the rated voltage. This particular case would result in a working capacitance of 42% (0.9 × 0.85 × 0.55) of the nominal value.

Select high-K ceramic capacitors according to the following rules:

1. Use capacitors made of materials with temperature coefficients of X5R, X7R, or better.
2. Use capacitors with DC-voltage ratings of at least twice the application voltage. Use minimum 10-V capacitors for the TPA2013D1.
3. Choose a capacitance value at least twice the nominal value calculated for the application. Multiply the nominal value by a factor of 2 for safety. If a 10-μF capacitor is required, use 20 μF.

The preceding rules and recommendations apply to capacitors used in connection with the TPA2013D1. The TPA2013D1 cannot meet its performance specifications if the rules and recommendations are not followed.

10.2.1.2.3.2 TPA2013D1 Capacitor Equations

The value of the boost capacitor is determined by the minimum value of working capacitance required for stability and the maximum voltage ripple allowed on V_CC in the application. The minimum value of working capacitance is 10 μF. Do not use any component with a working capacitance less than 10 μF.

For X5R or X7R ceramic capacitors, Equation 4 shows the relationship between the boost capacitance, C, to load current, load voltage, ripple voltage, input voltage, and switching frequency (I_CC, V_CC, ΔV, V_DD, f_boost respectively). Insert the maximum allowed ripple voltage into Equation 4 to solve for C. A factor of 2 is included to implement the rules and specifications listed earlier.

Equation 4.

For aluminum or tantalum capacitors, Equation 5 shows the relationship between the boost capacitance, C, to load current, load voltage, ripple voltage, input voltage, and switching frequency (I_CC, V_CC, ΔV, V_DD, f_boost respectively). Insert the maximum allowed ripple voltage into Equation 5 to solve for C. Solve this equation assuming ESR is zero.

Equation 5.

Table 7. Recommended Values

10.2.1.2.5.1 Decoupling Capacitors

The TPA2013D1 is a high-performance Class-D audio amplifier that requires adequate power-supply decoupling to ensure the efficiency is high and total harmonic distortion (THD) is low. Place a low equivalent-series-resistance (ESR) ceramic capacitor, typically 1 μF, as close as possible to the device VDD lead. This choice of capacitor and placement helps with higher frequency transients, spikes, or digital hash on the line. Additionally, placing this decoupling capacitor close to the TPA2013D1 is important for the efficiency of the Class-D amplifier, because any resistance or inductance in the trace between the device and the capacitor can cause a loss in efficiency. Place a capacitor of 10 μF or greater between the power supply and the boost inductor. The capacitor filters out high-frequency noise. More importantly, it acts as a charge reservoir, providing energy more quickly than the board supply, thus helping to prevent any droop.

10.2.1.2.5.2 Input Capacitors

The TPA2013D1 does not require input coupling capacitors if the design uses a differential source that is biased within the common mode input range. Use input coupling capacitors if the input signal is not biased within the recommended common-mode input range, if high-pass filtering is needed, or if using a single-ended source.

The input capacitors and input resistors form a high-pass filter with the corner frequency, f_c, determined in Equation 6.

Equation 6.

The value of the input capacitor is important to consider as it directly affects the bass (low frequency) performance of the circuit. Speakers in wireless phones cannot usually respond well to low frequencies, so the corner frequency can be set to block low frequencies in this application. Not using input capacitors can increase output offset.

Use Equation 7 to find the required the input coupling capacitance.

Equation 7.

Any mismatch in capacitance between the two inputs causes a mismatch in the corner frequencies. Choose capacitors with a tolerance of ±10% or better.