Spread spectrum modulation mode minimizes electromagnetic interference of Class D amplifiers-EEWORLD

Collect

Abstract: This article discusses two different Class D (switched mode) amplifier technologies: pulse width modulation (PWM) and spread spectrum modulation. Traditionally, PWM-based Class D amplifiers have required large and expensive filtering components to reduce electromagnetic interference (EMI) caused by their rail-to-rail switching and fast switching frequencies. Spread spectrum modulation techniques used in today's Class D amplifiers allow designers to eliminate these filtering components without sacrificing audio performance or amplifier efficiency.

introduction

Class D amplifiers are more attractive to designers of portable audio applications because they are more efficient than Class AB amplifiers. However, some designers do not use Class D amplifiers in portable applications because traditional PWM-based Class D amplifiers require large and expensive filtering components to reduce electromagnetic interference. Maxim's Class D amplifier spread-spectrum modulation technology allows designers to eliminate these filtering components without compromising audio performance or amplifier efficiency, thus effectively promoting the promotion of high-efficiency Class D amplifiers in portable audio applications.

Traditional PWM Amplifier Topology

Figure 1 shows a typical PWM-based, bridge-tied load (BTL) Class D amplifier. PWM schemes typically use an internally generated sawtooth waveform as a reference for their input stage. A comparator monitors the analog input voltage and compares it to the sawtooth waveform. When the sawtooth input exceeds the input voltage, the comparator output goes low. An inverter is used at the comparator output to generate a complementary PWM waveform that controls the second leg of the BTL output.

PWM-based amplifiers typically require bulky filtering components at their outputs because of their rail-to-rail switching characteristics and fast switching frequencies, which generate high radio frequency (RF) emissions and interference. An LC filter is typically required to reduce this high-frequency interference and extract the audio content from the duty cycle information of the PWM signal.

Figure 1. Traditional Pulse Width Modulation Topology

Spread Spectrum Modulation Amplifier Topology

One way to replace this expensive, large LC filter solution is to improve the switching process so that the amplifier can maintain high efficiency while reducing EMI. Maxim's Class D amplifiers do just that. This Class D amplifier uses a unique, patented spread-spectrum modulation scheme to broaden the broadband spectral components, thereby minimizing the EMI radiated by speakers and cables. Figure 2 shows this Class D amplifier topology using Maxim's MAX9700.

Maxim's Class D amplifier modulation scheme uses an internally generated sawtooth waveform and a complementary signal pair at the input. If the complementary input signal is not available, a differential input is generated inside the IC.

Figure 2. Mono Class D Amplifier Topology

The comparator monitors the input of the Class D amplifier and compares the complementary input voltage to the sawtooth waveform. When the amplitude of the sawtooth waveform exceeds the input voltage, comparator A outputs a low level, pulling the corresponding Class D output (OUT+) high to VDD. When the amplitude of the sawtooth waveform exceeds its input voltage, comparator B also outputs a low level, also pulling the corresponding Class D output (OUT-) high to VDD. After both Class D outputs are pulled high, a timer at the output of the NOR gate begins counting with a time constant tau, which is equal to 1 / (RTON * CTON). After the fixed time (tau) expires, both Class D outputs are pulled low to GND and both comparators are reset. This process produces a minimum pulse width tON (MIN) at the output of the second comparator. As the input voltage increases or decreases, the pulse duration of one of the outputs (which triggers the first comparator to flip) increases, while the pulse duration of the other output remains at tON(MIN), resulting in a change in the net voltage across the speaker (VOUT+ - VOUT-).

Figure 3. Output of a Maxim Class D BTL amplifier in FFM mode after an input signal is applied.

Figure 3. Output of a Maxim Class D BTL amplifier in FFM mode after an input signal is applied.

Figure 3. Output of Maxim's Class D BTL amplifier in FFM mode with an input signal

Fixed frequency modulation and spread spectrum modulation

Maxim's Class D amplifiers use two modulation modes: (1) fixed frequency modulation (FFM) mode and (2) spread-spectrum modulation mode. In FFM mode (Figure 3), the sawtooth period remains constant, as in conventional PWM schemes. In spread-spectrum modulation mode (Figure 4), the sawtooth period varies from cycle to cycle (by ±10%). The sawtooth period variation is exaggerated in Figure 4 to better illustrate the effect. Figure 4. Output of

Figure 4. Output of a Maxim Class D BTL amplifier in spread-spectrum modulation mode after the input signal is applied.

Figure 4. Output of a Maxim Class D BTL amplifier in spread-spectrum modulation mode after the input signal is applied.

Maxim's Class D BTL amplifier with input signal in spread-spectrum modulation mode.

In spread-spectrum modulation mode, the cycle-to-cycle variation of the period reduces the spectral energy at the fundamental frequency (fo ±10%) while expanding the harmonic content within a specific bandwidth (nfo ±10%, where n is a positive integer). Instead of concentrating a large amount of the spectral energy at multiples of the switching frequency, the energy is spread over a bandwidth that increases with frequency. Above several megahertz, the broadband spectrum looks like white noise, thus reducing EMI. In FFM mode, the energy is contained in a narrower frequency band and has a higher peak value (Figure 5a). In spread-spectrum modulation mode, the energy is contained in a wider frequency band and the peak energy is reduced (Figure 5b). Note that the third harmonic in Figure 5b is almost obscured by the noise floor.

Figure 5a. Maxim's FFM mode.

Figure 5a. Maxim's FFM mode

Figure 5b. Maxim's spread-spectrum modulation scheme.

Figure 5b. Maxim's spread-spectrum modulation mode

Spread spectrum modulation mode minimizes EMI emissions

Maxim's spread-spectrum modulation technology allows Class D amplifiers to be truly "filter-free" as long as the speaker cables are not too long. Conventional PWM architectures often require large output LC filters to ensure that consumer products using Class D amplifiers meet EMI regulations. Maxim's proprietary spread-spectrum modulation technology reduces the radiated emissions of Class D amplifiers, so that output filtering is not required or only minimal filtering components are required to meet EMI regulations (see Appendix).

EMI regulations require that end products pass existing quasi-peak detection limits—such as those set by the CE (European Community, European standards) and the FCC (Federal Communications Commission, American standards) to ensure the lowest levels of electromagnetic interference. As defined by these agencies, electromagnetic interference interrupts, hinders, or degrades the effective performance of electronic and/or electrical equipment. In quasi-peak detection, the measured signal level is measured by the repetition frequency of the signal's spectral components. The lower the repetition frequency, the lower the quasi-peak reading.

¹Spread-spectrum modulation takes advantage of the averaging nature of quasi-peak detection to significantly reduce EMI measurements (Table 1). In spread-spectrum modulation mode, the peak fundamental frequency of the Class D amplifier varies randomly within a certain range—usually within ±10% of its fundamental switching frequency. Assuming the analyzer is using 120kHz resolution bandwidth for quasi-peak detection, then the switching energy is present only for a short time at any single center frequency, in addition to the switching frequency fundamental and a few higher harmonics.

Table 1. Radiated emissions data for the MAX9759 (MAX9759EVKIT, using 3" twisted-pair speaker cable and "no filter" in spread-spectrum modulation mode) Table 1

in conclusion

The near-rail-to-rail switching characteristics and fast switching frequencies of Class D amplifiers can generate strong RF radiation and interference. Traditionally, large and expensive LC filters have been required to reduce this high-frequency interference before the audio content is reproduced by a sound-generating device such as a speaker. But now, with effective PCB layout and short speaker cables, Maxim's spread-spectrum modulation technology can achieve true "filterless" operation for low-power applications.

¹For more information on quasi-peak detection, see Reference Publication 16, published by the International Special Committee on Radio Interference (CISPR), a division of the International Electrotechnical Commission.

appendix

Filter Topology Overview

There are three filter topologies for Class D power amplifiers: (1) FB-C, ferrite beads and capacitors; (2) LC, inductors and capacitors; and (3) "no filter." Which filtering technology should be chosen for a particular design depends on the application's speaker cable length and PCB layout. Here are the pros and cons of each of the three filter topologies:

FB-C Filter

If the speaker cable length is moderate, FB-C filtering is sufficient to meet EMI limits. Compared with LC filtering, FB-C filtering is more compact and cost-effective. However, since it can only be effective at frequencies greater than 10MHz, the application range of FB-C filtering is greatly limited. Moreover, at frequencies below 10MHz, if the speaker cable routing is unreasonable, it will also cause conducted radiation to exceed the limit.

LC Filter

In contrast, LC filtering can start to suppress the effect at a frequency of about 30kHz. When the cables used in a design are long and the PCB layout is not very good, LC filtering is undoubtedly a "safe" choice. However, LC filtering requires expensive and bulky external components, which is obviously not suitable for portable devices. Moreover, when the frequency is greater than 30MHz, the main inductor will self-resonate, and additional components will be required to suppress electromagnetic interference.

"No filter" filtering

The "filterless" amplifier topology is the most cost-effective solution because it eliminates the need for additional filtering components. Class D amplifiers can meet EMC standards when short twisted-pair speaker cables are used. However, as with FB-C filtering, conducted emissions may be exceeded if the speaker cables are not routed properly. Also note that Maxim's Class D amplifiers can also achieve "filterless" operation as long as the speaker load is inductive at the amplifier's switching frequency. The large inductance at the switching frequency keeps the overload current relatively constant while the output voltage switches.

Reference address：Spread spectrum modulation mode minimizes electromagnetic interference of Class D amplifiers

Previous article：Reducing RF Susceptibility in Cell Phone Headphone Amplifiers
Next article：IF amplifier circuit diagram

Popular Resources
Popular amplifiers