When to Use An Output Filter

C2

C1

L1

L2

OUT−

OUT+

Gain Setting Via GAIN0 and GAIN1 Inputs

TPA2000D1-Q1

damage could occur to the speaker if the voice coil is not designed to handle the additional power. To size the

speaker for added power, the ripple current dissipated in the load needs to be calculated by subtracting the

minus the theoretical efficiency (

η

(1)

(2)

(3)

The maximum efficiency of the TPA2000D1 with an 8-

Ω load is 85%. Using Equation 3 with the efficiency at

maximum power (78%), we see that there is an additional 106 mW dissipated in the speaker. The added power

dissipated in the speaker is not an issue as long as it is taken into account when choosing the speaker.

Design the TPA2000D1 without the filter if the traces from amplifier to speaker are short. The TPA2000D1

passed FCC and CE radiated emissions with no shielding with speaker wires eight inches long or less. Notebook

PCs and powered speakers where the speaker is in the same enclosure as the amplifier are good applications

for class-D without a filter.

A ferrite bead filter (shown in Figure 11) often can be used if the design is failing radiated emissions without a

filter, and the frequency sensitive circuit is greater than 1 MHz. This is good for circuits that have to pass only

FCC and CE because FCC and CE test only radiated emissions greater than 30 MHz. If choosing a ferrite bead,

choose one with high impedance at high frequencies, but low impedance at low frequencies.

Use an output filter if the EMI sensitive circuits are low frequency (<1 MHz) and/or the leads from amplifier to

speaker are long.

The LC output filter is shown in Figure 11.

• L1 = L2 = 22 µH (DCR = 110 mΩ, part number = SCD0703T-220 M-S, manufacturer = GCI)

• C1 = C2 = 1 µF

The ferrite filter is shown in Figure 11, where L is a ferrite bead.

• L1 = L2 = ferrite bead (part number = MPZ1608S221, manufacturer = TDKI)

• C1 = C2 = 1 nF

Figure 11. Class-D Output Filter

The gain of the TPA2000D1 is set by two input terminals, GAIN0 and GAIN1.

The gains listed in Table 2 are realized by changing the taps on the input resistors inside the amplifier. This

ratios of resistors, so the actual gain distribution from part-to-part is quite good. However, the input impedance

may shift by 30% due to shifts in the actual resistance of the input resistors.

For design purposes, the input network (discussed in the next section) should be designed assuming an input

impedance of 20 k

Ω, which is the absolute minimum input impedance of the TPA2000D1. At the higher gain

settings, the input impedance could increase as high as 115 k

Ω.

Copyright © 2002–2008, Texas Instruments Incorporated

Submit Documentation Feedback

11

Product Folder Link(s): TPA2000D1-Q1