AN726_技术文档

AN726

UTILITY CLASS-D TOOLSTICK USER’S GUIDE

1. Introduction

Class-D is a switching power amplifier architecture that uses high-frequency Pulse-Width Modulation (PWM) to

generate the output waveform. The transistors are fully on or fully off, which means this architecture can reach very

high efficiency levels, resulting in significant power savings. However, the translation from the input signal to PWM

and the PWM itself can cause more distortion on the output than other amplifier architectures. Figure 1 shows a

very basic Class-D signal diagram.

Figure 1. Class-D Power Amplifier Signal Diagram

Because the Class-D amplifier transistors are fully on or fully off, the transistors lose less energy in the form of

heat. This architecture, therefore, lends itself to small, cheap MOSFETs. The Class-D amplifier can be

implemented in either analog or digital form.

An analog Class-D amplifier is non-trivial and typically consists of a comparator, triangle waveform generator, and

several blocks to condition the input signal before transferring to the output MOSFETs.

A digital Class-D amplifier requires the following:

PWM output (switching) frequency 10x+ faster than the highest input frequency to adequately reconstruct

the input signal

High-resolution control of the PWM pulse width to reduce output quantization distortion

Method for sampling or receiving the input waveform

Fast core for digital processing and manipulation of data

Pins capable of driving the amplified signal

Since the Silicon Labs Precision32™ SiM3U1xx/SiM3C1xx 32-bit MCUs have peripherals and features capable of

meeting all of these requirements, these MCUs are uniquely suited to a utility Class-D power amplifier application

without using many external components by directly driving the speaker using the high drive I/O.

In addition to providing a development platform for Class-D applications, the Class-D ToolStick board serves as a

lower-cost, general-purpose development platform for the SiM3U1xx USB MCUs.

2. Relevant Documents

This document provides a hardware and software overview for the Class-D ToolStick. Additional documentation on

the Precision32 tools and MCUs can be found on the Silicon Labs website: www.silabs.com/32bit-appnotes,

www.silabs.com/32bit-software, and www.silabs.com/32bit-mcu.

Rev. 0.1 11/12

AN726

3. Using the Demo

The speaker in the kit connects to the JP1 1x2 header next to the green terminal block (J4). Two alternate

connection methods are available to support a variety of speakers: the J4 terminal block, and an output audio jack

(J5). A screwdriver can be used to attach the speaker(s) of choice to J4, and each speaker should have two leads:

positive (red) and negative (black). The audio jack output connects both channels of the speaker to R+ and R–, so

it is not true mono. The Class-D ToolStick board is intended for use with stereo systems.

The Class-D ToolStick can be powered from either the Debug (J3) or Device (J2) USB mini connectors as shown in

Figure 2. The green POWER LED (DS7) will light up when the board is powered. Do not touch the Capacitive

Sensing slider during power-on, as this is when the board calibrates the slider. Once the board is ready for use, the

blue LED (DS1) will turn on.

Figure 2. Powering the Board

Change the volume of the output at any time by pressing firmly and sliding on the Capacitive Sensing slider (CS1).

The blue LEDs above the slider indicate the volume level.

The MODE button on the slider can be used to change the mode of the board:

Mode 1 (blue LED DS1): Stereo Jack Input Mode

Mode 2 (blue LED DS2): USB Input/Output Mode

Mode 3 (blue LED DS3): Play Prerecorded Flash Mode

Mode 4 (blue LED DS4): Play/Record Flash Mode

2

Rev. 0.1

AN726

3.1. Stereo Jack Input Mode

The default mode of the Class-D ToolStick is stereo jack input mode, indicated by blue LED DS1. To play sound in

this mode, connect any input source (MP3 player, phone, etc.) to the J1 stereo jack. The volume level of the input

device should be as high as possible without causing saturation, which the red LED (DS8) indicates.

3.2. USB Input/Output Mode

In USB mode, the board plays sound from the USB host PC or records sound from the board microphone (MK1)

and sends it up to the PC. To use the board in this mode, connect the Device USB connector (J2) to the host PC

and press the MODE button until blue LED DS2 turns on. The board should appear as the Utility Class-D

ToolStick in Device Manager once it’s properly enumerated as a USB Audio device.

To select the Class-D Audio device as the Playback or Recording source in Windows 7, right-click on the Volume

Mixer icon in the System Tray and select either Playback devices or Recording devices. In the Playback tab of

the dialog, select the Utility Class-D ToolStick and press the Set Default button as shown in Figure 3. The

settings for the device can be selected using the Properties button in this tab.

Figure 3. Configuring the USB Utility Class-D ToolStick as the Playback or Recording Device

In the Recording tab of the dialog, select the Utility Class-D ToolStick microphone and press the Set Default

button. This will open an interface to the microphone and will display the recorded input volume to the right of the

device name. If this volume is not registering sounds, press the Properties button to configure the microphone

volume.

Once configured, any program on the PC should output to the Class-D ToolStick device. Ensure that the volume

levels in the program and Volume Mixer are set to appropriate levels. If the device volume settings are not affecting

the Class-D board, click the expansion button below the Speakers icon and select the Speakers (Utility Class-D

ToolStick) as shown in Figure 4.

When recording, the red LED (DS8) indicates when the microphone input saturates.

Note: As long as a recording interface is open (i.e. Volume Mixer Recording devices dialog or a recording program open), the

Class-D ToolStick prevents the user from switching the operation mode. This is due to potential synchronization issues

when switching between different modes and sending data up to the host PC. To switch the mode, close the Volume

Mixer or the recording program.

Rev. 0.1

3

AN726

Figure 4. Selecting the USB Utility Class-D ToolStick in Volume Mixer

3.3. Play Prerecorded Flash Mode

In prerecorded flash mode, the Class-D ToolStick will output ~7 seconds of sound from an array saved to flash.

Enter the mode by pressing the MODE button until blue LED DS3 turns on.

Press the PLAY slider button on the board to start the waveform, and press the PLAY slider button again to stop it

before it completes. The blue LED (DS6) will indicate when a play operation is in progress. This mode provides an

alternative sound input source when no sources are available and provides an indication of what sound effects

would sound like.

3.4. Play/Record Flash Mode

The play/record flash mode records compressed sound to the flash of the MCU. Enter the mode by pressing the

MODE button until blue LED DS4 turns on.

Press the REC slider button to start recording, and press the REC slider button again to stop recording. The blue

LED (DS5) will indicate when a record operation is in progress. Multiple recordings will append to the full recording

in flash. Once the flash is full, the MCU will indicate no additional recordings can be made by turning on and

immediately turning off the blue record LED. To record additional sound, press the ERASE slider button to erase

the flash. The blue DS5 LED will turn on while the erase operation is in progress. When recording, the red LED

(DS8) indicates when the microphone input saturates.

Press the PLAY slider button on the board to start the waveform, and press the PLAY slider button again to stop it

before it completes. The blue LED (DS6) will indicate when a play operation is in progress.

4

Rev. 0.1

AN726

4. Class-D ToolStick Hardware Overview

The Class-D ToolStick board enables Class-D application development on the SiM3U164 MCU. In addition to

providing a development platform for Class-D applications, the Class-D ToolStick board serves as a low-cost,

general-purpose development platform for the SiM3U1xx USB MCUs.

Figure 5 shows the Utility Class-D ToolStick board features. Full schematics for the board can be found in

7. "Schematics‚" on page 22

Testpoint Pin Access

for General-Purpose

Development

ToolStick Debug

Adapter

Class-D Output

Filters and

Speaker Header

Optional High

Power Components

Audio Inputs

Capacitive Sensing

Stereo Jack and

Slider and LED User

Interface

Microphone Gain/Bias

Figure 5. Utility Class-D ToolStick Board Features

4.1. Push-Button Switches and LEDs (S1-2, DS1-DS8)

The Class-D ToolStick board has two push-button switches and eight LEDs summarized in Table 1. The switches

connect to PB0.6 (S1) and PB0.7 (S2) and are currently unused in the Class-D firmware. The switches are

normally open and pull the pin voltage to ground when pressed.

Port pins PB0.1, PB0.2, PB0.4, PB0.5, PB0.10, and PB0.11 connect to six blue LEDs (DS1-DS6). Pin PB3.3

connects to the red LED (DS8). The green LED, POWER (DS7), turns on when USB power is applied to the board

from either USB connector. The LEDs connect to VIO through a current limiting resistor.

Table 1. Class-D ToolStick Switches and LEDs

GPIO Pin

PB0.6

PB0.7

PB0.1

PB0.2

PB0.4

Switch or LED

Push-Button Switch (S1)

Push-Button Switch (S2)

Blue LED (DS1)

Blue LED (DS2)

Blue LED (DS3)

Rev. 0.1

5

AN726

Table 1. Class-D ToolStick Switches and LEDs (Continued)

GPIO Pin

PB0.5

PB0.10

PB0.11

PB3.3

—

Switch or LED

Blue LED (DS4)

Blue LED (DS5)

Blue LED (DS6)

Red LED (DS8)

Green POWER LED (DS7)

4.2. Class-D Output Network, Terminal (J4), Header (JP1), and Jack (J5)

The four high drive I/O pins on the SiM3U164 40-pin QFN package connect to the J4 terminals after going through

the Class-D output network, allowing these pins to directly drive the speaker(s). The JP1 1x2 header connects to

R+ and R– to interface with the kit speaker. The output jack (J5) connects R+ and R– to both channels and does

not provide true mono sound. To create a mono output, connect the L_IN and R_IN test points together in the

stereo jack and microphone gain/bias section of the board.

The output network consists of a ferrite bead, LC, and grounding caps for each pair of channels, R+/R– and L+/L–.

These components should be tuned for the specific application speaker as discussed in 6. "Tuning the Class-D

Output Network‚" on page 18.

Table 2. Terminal Block Pin Descriptions (J4)

1

I/O

PB4.0 / R+

PB4.1 / R–

PB4.2 / L+

PB4.3 / L–

2

3

4

4.3. ToolStick Debug Adapter (U3)

The Class-D ToolStick board features a debug adapter via the mini-B USB connector (J3) labeled DEBUG. This

debug adapter can be used with the Precision32 (1.0.2 and higher) and ARM uVision IDEs (4.54 and higher).

When using an older version of the IDEs, the debug adapter DLLs in the IDE directories must be replaced to

support the ToolStick Debug Adapter. Contact technical support (“Contact Information” section) for more

information.

4.4. USB Connection Options (J2)

The Class-D ToolStick board features a USB connection via the mini-B USB connector (J2) labeled DEVICE. This

USB connector provides the USB Audio features of the board and facilitates general-purpose USB development on

the SiM3U164 device.

6

Rev. 0.1

AN726

4.5. Capacitive Sensing Buttons and Slider (CS1)

Four capacitive sensing buttons and one slider comprised of six segments are included on the board to provide the

user interface. Six GPIO pins are connected to the slider as shown in Table 3. The GPIO pins are connected to the

slider in descending order from the location labeled MODE to the location labeled PLAY on the board.

The test points above the slider facilitate adding grounded or shrouded connectors to measure the capacitive

sensing pins.

Table 3. Capacitive Sensing GPIO Connections

Location

MODE

—

PB1.1

PB1.0

ERASE

—

PB0.15

PB0.14

PB0.13

PB0.12

REC

PLAY

4.6. Analog Audio Inputs (J1 and MK1)

The stereo jack on the board (J1) allows the board to measure stereo inputs from an input source. The microphone

(MK1) enables the device to measure voice data directly. These signals are conditioned by a bias and gain circuit

provided by a quad op-amp IC (U1), resistors, and capacitors to a range appropriate for the SiM3U164 SARADC

modules.

Rev. 0.1

7

AN726

5. Class-D ToolStick Firmware Overview

5.1. System Overview

The system is configured for 48 MHz AHB and APB to maximize the output resolution of the EPCA.

In all modes, the EPCA0 operates continuously and outputs the PWM signal to the speaker. This waveform

generation is never interrupted. The SARADC modules operate in 12-bit mode at the maximum SAR clock

frequency in order to complete conversions as quickly as possible. The USB internal oscillator provides the

48 MHz system clock, which in conjunction with the PLL0 module, provides the precise timing required in the

different modes to recreate the audio waveforms.

5.1.1. SiM3U164 Module Usage

The modules used by the Class-D ToolStick firmware are:

SARADC0: Measures the right stereo jack channel and the microphone input

SARADC1: Measures the left stereo jack channel

VREF0: Provides the 2.4 V reference for the ADCs

FLASHCTRL0: Writes recorded microphone data to flash

EPCA0: Generates Class-D PWM waveforms (four channels)

CAPSENSE0: Provides button and slider user interface

TIMER0: Generates ~20 ms capacitive sensing scan timer

TIMER1: Generates mode timing (e.g. 48 kHz in USB) and handles LED PWM updates

PCA0 and Crossbar: Generates LED PWM waveforms for the six blue LEDs (one channel)

USB0: Provides the USB Audio input/output interface

PLL0: Provides the precise timing for the system and spectrum spreading to reduce emissions

AppBuilder generates the initialization code for all modules except for the USB0 module. This AppBuilder project is

included in the Utility Class-D ToolStick software package.

Note: Exporting code from version 1.1.1 of AppBuilder will overwrite the project file associated with the Class-D ToolStick firm-

ware. This project file includes manual changes to include the si32Library that will be removed when AppBuilder exports

the code. A saved version of the project file (save.project) is available in the Class-D ToolStick firmware package in

case this project file is overwritten. To restore the project file, simply copy the contents of save.project to the .project

file.

5.1.2. Firmware Organization

The Class-D ToolStick code follows the AppBuilder code organization scheme. The application-specific code for

each peripheral is in the src folder in the my- files, and the generated files from AppBuilder are in a generated

folder below the src folder. Additional application-specific files are in the src folder with names associated with the

functions in the file, like mulaw or LED_control.

class_d: includes the main Class-D loop and any control not handled by the TIMER1L interrupt handler.

LED_control: blue and red LED update functions; handles the shifting of the PCA channel on the crossbar.

mulaw: implements the µLaw companding algorithms.

myApplication: implements the system initialization code before passing control to class_d.

myBuildOptions: required for si32Library.

myDataPlaneIn: si32Library callback functions for the USB IN transactions (recording).

myDataPlaneOut: si32Library callback functions for the USB OUT transactions (playing).

myUsbAudioDevice: USB Audio descriptors and audio-specific si32Library function calls.

myUSB0: non-audio USB si32Library functions calls for the application.

pre_recorded_array: µLaw compressed prerecorded flash data.

volume: handles the volume decoding and control from either the USB interface or the Capacitive Sensing

slider.

my- files: implement application code associated with the particular peripheral.

g- files: AppBuilder generated initialization code associated with the particular peripheral.

8

Rev. 0.1

AN726

5.2. Algorithms

The Class-D ToolStick uses two algorithms to manipulate data: Exponentially Weighted Moving Average (EWMA)

and µLaw. This section also describes Remainder-Weighted Dithering as an optional algorithm.

5.2.1. Exponentially Weighted Moving Average (EWMA)

The EWMA is a form of Exponential Averaging that creates a low-pass filter. This filter performs noise shaping to

help move the quantization noise out of the audible range. The Class-D ToolStick firmware uses this form of

averaging to filter the ADC results.

The EWMA takes the form:

yn – 1

xn

--------------------

----------

yn = yn – 1 –

+

2^m

In this equation, y(n) is the current output, y(n-1) is the previous output, x(n) is the current input sample, and m is

the weight of the average.

The weight determines how heavy the average of the EWMA is. The smaller weights (m = 1, 2) use more of the

new sample and react more quickly to dramatic changes in the input. Larger weights (m = 4, 5) use more of the

previous output and provide more smoothing.

The EWMA has several advantages over other forms of averaging:

Quick to calculate (two shifts, an addition, and a subtraction)

Only saving the previous value

Contains history from every sample in time, which would take lots of memory in a traditional average

The weight can be dynamically adjusted easily in firmware depending on the circumstances without

dramatically changing the computational time

The EWMA is a form of IIR filter, which means it can become unstable if the cumulative error becomes too large.

5.2.2. Remainder-Weighted Dithering

The firmware quantizes the 12-bit ADC results or 16-bit signed USB data to a 9-bit EPCA PWM value. Remainder-

weighted dithering is an optional algorithm that allows the Class-D ToolStick firmware to recover some of the

energy lost in the LSBs during this conversion in a random, inaudible way.

This algorithm captures the LSBs of the original 12- or 16-bit data, shifts the data to the final 9-bit size, and

compares against a random value of the same magnitude as the LSBs shifted away. If the original data was

positive and the LSBs are greater than the random value, then the LSB of the 9-bit data is incremented by 1.

Similarly, if the original data was negative and the LSBs are less than the random value, the LSB is decremented

by 1. Otherwise, the data remains unchanged.

12-bit to 9-bit (3 bits):

output_value = 1931 = 0x078B

output_lsb = 0x78B & 0x7 = 3

output_value = 1931 >> 3 = 241

random_lsb = 1

random_lsb = 6

random_lsb = 4

random_lsb = 0

random_lsb = 3

242

241

EPCA

overflows

Figure 6. Remainder-Weighted Dithering Example

Rev. 0.1

9

AN726

The example in Figure 6 assumes a original 12-bit signed data value of 1931 and shows how the average energy

increases. In this particular example, the data remains the same for each EPCA overflow for simplicity, but this will

not necessarily be the case in the operation of the algorithm.

5.2.3. µLaw (or muLaw)

The µLaw companding algorithm has long been used in telephony and other voice applications. This algorithm

takes advantage of the behavior of the human ear by reserving most of the compression bins for low volume levels

where the ear is most sensitive. Using this algorithm, the Class-D firmware can store more data in flash with no

noticeable degradation in audio quality.

This algorithm takes a 14-bit signed number and adds 32 to the magnitude, which ensures that a 1 occurs in bits 5–

12 of the value. This means the valid input range is –8160 to 8159. This value is then converted to an 8-bit

compressed result as shown in Figure 7, where S is the sign bit. Finally, the 8-bit value is complemented.

14-bit Signed Input Data

8-bit µLaw Encoded Data

S 0 0 0 A B C D

S 0 0 1 A B C D

S 0 1 0 A B C D

S 0 1 1 A B C D

S 1 0 0 A B C D

S 1 0 1 A B C D

S 1 1 0 A B C D

S 1 1 1 A B C D

S 0 0 0 0 0 0 0 1 A B C D X

S 0 0 0 0 0 0 1 A B C D X X

S 0 0 0 0 0 1 A B C D X X X

S 0 0 0 0 1 A B C D X X X X

S 0 0 0 1 A B C D X X X X X

S 0 0 1 A B C D X X X X X X

S 0 1 A B C D X X X X X X X

S 1 A B C D X X X X X X X X

Figure 7. µLaw Algorithm Table

Figure 8 shows the plot of the 14-bit signed inputs versus the 8-bit compressed output.

100

50

Input Data

-8000

-6000

-4000

-2000

2000

4000

6000

8000

-50

-100

Compressed

Output

Figure 8. µLaw Algorithm Plot

10

Rev. 0.1

AN726

5.3. EPCA0 Class-D Output

In all modes, the EPCA0 module runs in center-aligned PWM mode with an upper limit of 256. This means that the

channels operate in the range of 0–511, since the channels count in half clocks. When the positive channel is on

longer than the negative channel, the resulting equivalent energy is positive (above zero). When the negative

channel is on longer than the positive channel, the resulting equivalent energy is negative (below zero). Combined

with the components on the output, this PWM scheme generates the output waveform on the speaker as shown in

Figure 9.

FB

L

PB4.0 (R+)

C_g

C

FB

L

PB4.1 (R-)

C_g

CH0 (R+)

CH1 (R-)

Differential

Figure 9. Class-D Output Example

Note: The output network configuration in Figure 9 is the configuration in Revision 1.0 of the hardware. An alternate output net-

work discussed in 6. "Tuning the Class-D Output Network‚" on page 18 may provide better performance than the original

configuration.

5.4. Volume Control

The volume control on the Class-D ToolStick board uses the current-limiting feature of the high drive I/O to adjust

the output power. This has the additional benefit of allowing the PWM to always operate at full range, even at lower

volumes, which reduces firmware size and improves sound quality.

Rev. 0.1

11

AN726

5.5. Stereo Jack Input Mode

In stereo jack input mode, the goal is to sample at an even division of the EPCA0 update frequency to provide the

lowest distortion attainable. The data from the ADCs is 12-bit unsigned.

Class-D RD Board

R

SiM3U1xx

SARADC0

SARADC1

EPCA

L

Figure 10. Stereo Jack Input Mode Block Diagram

This mode measures the input source from the stereo jack (J1) using both SARADC modules, performs the EWMA

algorithm, quantizes the data to 9 bits, and outputs the PWM waveform on the EPCA0 module. Figure 11 shows

the data flow for the stereo input jack mode.

The Class-D ToolStick board automatically biases the input waveform to VREF/2. On hardware that does not do

this, an additional dc offset adjustment algorithm may be required.

R

SiM3U1xx

hardware

SARADC0

gain stage

adjusting

DC offset

algorithm

remainder-

weighted

dither

EPCA

hardware

SARADC1

gain stage

9-bit

quantization

EWMA

L

Figure 11. Stereo Jack Input Mode Flow Diagram

12

Rev. 0.1

AN726

5.6. USB Input/Output Mode

In USB input/output mode, the system samples and outputs data at 48 kHz. The USB data is 16-bit signed data.

Class-D RD Board

R

SiM3U1xx

Host PC

USB

sound

SARADC0

software

MIC

EPCA

L

Figure 12. USB Input/Output Mode Block Diagram

For playing (from PC to the board), the board receives data from the USB interface using the si32Library USB

Audio Component and outputs the PWM waveform on the EPCA0 module.

For recording (from board to the PC), the firmware measures the microphone (MK1) using one SARADC module

and sends the data to the PC using the si32Library USB Audio Component.

Figure 13 shows the data flow for the USB input/output mode.

SiM3U1xx

Host PC

R

USB0

9-bit

sound

quantization

software

shift from 12

bits unsigned

to 16 signed

remainder-

weighted

dither

hardware

gain stage

EPCA

SARADC0

MIC

L

Figure 13. USB Input/Output Mode Flow Diagram

Rev. 0.1

13

AN726

5.7. Play Prerecorded Flash Mode

In play prerecorded flash mode, the system outputs data from the flash at 9.6 kHz.

R

Class-D RD Board

SiM3U1xx

Flash

EPCA

L

Figure 14. Play Prerecorded Flash Mode Block Diagram

The firmware reads the compressed data from the flash, decompresses the data using µLaw, performs the EWMA

algorithm, quantizes the data to 9 bits, and outputs the PWM waveform on the EPCA0 module. The remainder-

weighted dithering algorithm is not used at all in this mode since the data LSBs will often be 0s from the

compression algorithm.

Figure 15 shows the data flow for the play prerecorded flash mode.

R

SiM3U1xx

µLaw

9-bit

decompression 8

bits to 14 (signed)

Flash

EPCA

quantization

12-bit

quantization

EWMA

L

Figure 15. Play Prerecorded Flash Mode Flow Diagram

14

Rev. 0.1

AN726

5.8. Play/Record Flash Mode

In play/record flash mode, the system samples and outputs data at 9.6 kHz.

Class-D RD Board

R

SiM3U1xx

Flash

SARADC0

MIC

EPCA

L

Figure 16. Play/Record Flash Mode Block Diagram

For playing, the firmware reads the compressed data from the flash, decompresses the data using µLaw, performs

the EWMA algorithm, quantizes the data to 9 bits, and outputs the PWM waveform on the EPCA0 module. The

remainder-weighted dithering algorithm is not used at all in this mode since the data LSBs will often be 0’s from the

compression algorithm.

For recording, the firmware measures the microphone (MK1) using one SARADC module, compresses the data

using µLaw, and stores the data in flash.

Figure 17 shows the data flow for the play/record flash mode.

R

SiM3U1xx

µLaw

9-bit

decompression 8

bits to 14 (signed)

Flash

EPCA

quantization

µLaw compression

14 bits to 8 (signed)

12-bit

quantization

EWMA

L

convert from 12

bits unsigned to

14 signed

hardware

gain stage

SARADC0

MIC

Figure 17. Play/Record Flash Mode Flow Diagram

Rev. 0.1

15

AN726

5.9. Build Configurations

The Precision32 and uVision projects include two build configurations each: Release and Debug. These

configurations change the code optimization settings. In addition, each configuration sets a flag to allow the

firmware to automatically detect the current build configuration and IDE. This enables the firmware to modify the

play/record flash mode array size to appropriate settings based on the code size. The approximate sizes for each

build configuration are shown in Table 4.

Table 4. Build Code Size Comparisons

IDE

Build

Debug

Code Size (bytes)

68200

Precision32

Release

Debug

47200

uVision4

40200

Release

34700

To change the active build configuration in the Precision32 IDE, right-click on the project name in the Project view

and select Build ConfigurationsSet ActiveDebug or Release.

Figure 18. Setting the Active Precision32 IDE Build Configuration

16

Rev. 0.1

AN726

To change the active build configuration in uVision4, select the configuration from the drop-down menu shown in

Figure 19.

Figure 19. Setting the Active uVision4 IDE Build Configuration

5.10. si32 HAL Version

The Class-D ToolStick firmware was originally developed for v1.1.1 of the HAL or later. Compiling against previous

versions of the HAL will generate errors.

Rev. 0.1

17

AN726

6. Tuning the Class-D Output Network

6.1. Revision 1.0 Configuration

Class-D is an extension of a single-ended, single-pin speaker drive, where one side of the speaker is driven

directly (perhaps with a ferrite bead) and one side is grounded. This output method works very well with SiM3U1xx

devices.

The full Class-D output network consists of ferrite beads, inductor-capacitor (LC) filters, and grounding capacitors

(C ). Figure 20 shows the output network configuration used by the Revision 1.0 hardware. An alternate output

g

network shown in Figure 21 may provide better performance than the original configuration.

FB

L

PB4.0 (R+)

C_g

C

PB4.1 (R-)

C_g

Figure 20. Revision 1.0 Hardware Class-D Output Network—Direct Drive

The ferrite beads dissipate pad switching noise as heat for EMI purposes, the LC components filter the PWM noise,

and the grounding capacitors provide additional filtering in conjunction with the speaker reactance.

In almost all applications, the ferrite beads should be populated on the board. However, the other components may

or may not be populated, depending on the speaker used and the cost requirements of the application. If either the

L or the C are installed, both should be populated to ensure lower power consumption. In most cases, the easiest

method of determining whether to install the components and the values for these components is through physical

experimentation. The perceived quality can be determined by the following characteristics:

loudness

changes in loudness across bandwidth

bandwidth

noise harmonics across bandwidth

hiss at low and high volumes

Other factors to consider are the application’s sound source (i.e. USB data versus sampled ADC data) and the

maximum power consumption of the board in the application, which will determine how much data processing can

occur and if external MOSFETs are appropriate. The Class-D board provides 0  resistor pads to bypass the

series components, if desired.

18

Rev. 0.1

AN726

6.2. Alternate Configuration

Figure 20 shows the output network configuration used by the Revision 1.0 hardware. An alternate output network

shown in Figure 21 may provide better performance than the original configuration.

placed as close to the

speaker connectors as

device as possible

possible

L

FB

PB4.0 (R+)

C

C_g

C_B

PB4.1 (R-)

C_g

C

Figure 21. Alternate Class-D Output Network—Direct Drive

For the alternate configuration, the recommended components for the board are shown in Table 5.

Table 5. Class-D ToolStick Alternate Output Network Recommended Components

Reference

Value

~3.3 µH, 1.5 A

~2.2 µF

L

C

~20 pF

g

C

~0.47 µF

B

Rev. 0.1

19

AN726

6.3. Revision 1.0 High Power Configuration

Instead of directly driving the speaker, the high-drive I/O can interface with external MOSFETs as shown in

Figure 22, enabling higher-power applications. The board includes footprints for these external MOSFETs to

enable development with this configuration. The pull-up and pull-down resistors (~100 k) ensure that the tristate

reset state of the high drive I/O does not cause the FETs to accidentally conduct current. When using these

external MOSFETs, uninstall the R32, R35, R38, and R41 0  resistors.

Figure 23 illustrates an alternate full-bridge high-power output network. The diode and RC circuits on the MOSFET

gates ensure that the two devices will not both be conducting simultaneously.

FB

L

PB4.0 (R+)

C_g

C

VIOHD

FB

L

PB4.1 (R-)

C_g

Figure 22. Half-Bridge Class-D Output Network–External MOSFETs

The Revision 1.0 hardware provides footprints to a 9 V dc adapter connector and high power LDO. These

components can be populated to provide a higher current source to the external MOSFET transistors and VIOHD.

Alternatively, an external supply can connect to the transistors alone by removing the R48 0  resistor and placing

the positive terminal on the P_PWR test point.

20

Rev. 0.1

AN726

6.4. Alternate High Power Configuration

The Revision 1.0 hardware uses a half-bridge configuration shown in Figure 22. This configuration can lead to

higher distortion. Figure 23 illustrates an alternate full-bridge high-power output network. The diode and RC circuits

on the MOSFET gates ensure that the two devices will not both be conducting simultaneously. The value of these

components should be tuned for the gate capacitance of the FETs.

placed as close to the

drivers as possible

speaker connectors as

possible

VIOHD

L

FB

PB4.0 (R+)

C

C_g

C_B

VIOHD

L

FB

PB4.1 (R-)

C

C_g

Figure 23. Full-Bridge Class-D Output Network—External MOSFETs

Rev. 0.1

21

AN726

7. Schematics

4

1 1

22

Rev. 0.1

AN726

6

5

4

3

2

1

E P A D

2 9

S

G P I O / C T

G P I O / R T

O D / S T W O / R X

N C

2 2

2 3

2 4

2 5

2 6

2 7

2 8

1 4

1 3

L E D _ S T O P

1 2

T D I / T X

S / M S T W D I O

n S R S T

K / C S T W C L K

S U S P E N D

1 1

C 2 D

1 0

R S T b / C 2 C K

9

V B U S

P B 0 . 3

P B 0 . 2

P B 0 . 1

P B 0 . 0

P B 0 . 1 2

3 1

3 2

3 3

3 4

3 5

3 6

3 7

3 8

3 9

4 0

2 0

P B 0 . 1 3

1 9

P B 0 . 1 4

8

1 8

P B 0 . 1 5

1 7

V D D

P B 1 . 0

1 6

V R E G I N

V B U S

P B 1 . 1

1 5

V S S

1 4

D +

D -

V I O

1 3

P B 1 . 2

1 2

R E S E T

P B 1 . 3

1 1

4

S H

Rev. 0.1

23

AN726

24

Rev. 0.1

AN726

8. Bill of Materials

Table 6. Utility Class-D ToolStick Board Bill of Materials

Reference

C1, C3

Part Number

Source

Venkel

Description

C0603X5R6R3-106M

C0603X7R250-474K

C0603X7R100-105K

C1608X7R1H223M

10 µF, 6.3 V ±r% X5R 0603

0.47 µF, 25 V ±10% X7R 0603

1 µF, 10 V ±10% X7R 0603

0.022 µF, 50 V ±20% X7R 0603

C2, C6, C8

C24

Venkel

C26, C28, C29,

C31

TDK Corporation

C4, C10, C12,

C13, C15, C17,

C19, C20, C23,

C25

C0603X7R100-104K

Venkel

0.1 µF, 10 V ±10% X7R 0603

C5, C7

C0603X7R100-225K

C0603X5R6R3-475K

Venkel

2.2 µF, 10 V ±10% X7R 0603

4.7 µF, 6.3 V ±10% X5R 0603

C9, C11, C14,

C16, C18, C21,

C22

D1, D3

D4

SP0503BAHTG

RB550EATR

Littlefuse

USB Diodes 300 mW, 20 V TVS SOT143

Diode 700 mA, 30 V Dual Schottky SOT5N

Blue 15 mA, 2.85 V SMT, ChipLED 0603

Rohm Semiconductor

DS1, DS2,

DS3, DS4,

DS5, DS6

LB Q39G-L2N2-35-1

OSRAM Opto

Semiconductors Inc

DS7

DS8

SML-LX0603SUGW

SML-LX0603IW

LUMEX INC

GREEN 25 mA 0603

RED 30 mA 0603

FB1, FB2, FB3,

FB4

MPZ1608S601A

TDK Corporation

600 , 1 A SMT 0603

J1, J5

J2, J3

J4

SJ1-3543N

UX60-MB-5ST

CUI Inc

HIROSE

Phoenix Contact

Samtec

TDK

Stereo Audio Jack

USB mini connector B

1729144

1x4 Terminal Block

JP1

TSW-102-07-T-S

NLV25T-R68J-PF

84332C

1x2 Jumper Header

L1

0.68 µH, 300 mA ±5% GP

L2, L3, L4, L5

LS1

Murata

3.3 µH, 1.5 A ±20% Unshielded

Speaker 1 W Max

GA0571H

ALC

MK1

CMA-4544PF-W

CR0603-10W-2001F

CR0603-10W-1003F

CR0603-16W-2803F

CR0603-10W-1002F

CR0603-8W-5111F

CR0603-16W-21R0F

CR0603-10W-1001F

CUI Inc

Venkel

MICROPHONE MIC-2.5X9.7MM-RAD

2 k, 1/10 W ±1% ThickFilm 0603

100 k, 1/10 W ±1% ThickFilm 0603

280 k, 1/16 W ±1% ThickFilm 0603

10 k, 1/10 W ±1% ThickFilm 0603

5.11 k, 1/8 W ±1% ThickFilm 0603

21 , 1/16 W ±1% ThickFilm 0603

1 k, 1/10 W ±1% ThickFilm 0603

R1

R11, R18

R12, R19

R13

Venkel

R14

Venkel

R2, R7, R15

Venkel

R20, R29, R44,

R49

Venkel

R28, R42

CR0603-16W-7500F

Venkel

750 , 1/10 W ±1% ThickFilm 0603

Rev. 0.1

25

AN726

Table 6. Utility Class-D ToolStick Board Bill of Materials (Continued)

Reference

Part Number

CR0603-10W-4703F

CR0603-16W-000

Source

Venkel

Description

R3, R10, R17

470 k, 1/10 W ±1% ThickFilm 0603

0 , 1 A ThickFilm 0603

R31, R32, R34,

R35, R37, R38,

R40, R41, R48

R4

R5

R6

R8

CR0603-16W-2741F

CR0603-10W-2003F

CR0603-10W-2492F

CR0603-16W-4991F

CR0603-10W-1000F

Venkel

2.74 k, 1/16 W ±1% ThickFilm 0603

200 k, 1/10 W ±1% ThickFilm 0603

24.9 k, 1/10 W ±1% ThickFilm 0603

4.99 k, 1/16 W ±1% ThickFilm 0603

100 , 1/10 W ±1% ThickFilm 0603

R9, R16, R21,

R22, R23, R24,

R25, R26

S1, S2

EVQ-PAD04M

NSS-4-4-01

Panasonic Corp

Momentary Tactile Switch 6.5x4.5

4-40 HDW

SC1, SC2,

SC3, SC4

Richco Plastic Co

SO1, SO2,

SO3, SO4

2397

SPC Technology

STANDOFF HDW

U1

U2

U3

MC33204DR2

SiM3U164-B-GM

CF326-SX0261GM

ON Semiconductor

Silicon Laboratories

Quad Op-Amp

SiM3U164-B-GQ MCU QFN40 M6x6 P0.5

ToolStick Debug Adapter MCU

Components Not Installed

C27, C30

C0603X7R100-225K

Venkel

2.2 µF, 10 V ±10% X7R 0603 Not Installed

(NI)

C32

C0603X7R100-334K

C0603X5R6R3-475K

RAPC722X

0.33 µF, 10 V ±10% X7R 0603 Not Installed

(NI)

C33

Venkel

4.7 µF, 6.3 V ±10% X7R 0603 Not Installed

(NI)

J6

Switchcraft Inc.

ON Semiconductor

3-pin Power Jack 5A BARREL Not Installed

(NI)

M1, M2

NTR4171PT1G

–3.5 A, –30 V Pchan SOT23 Not Installed

(NI)

Q1, Q2

R27, R45, R50

R46, R47

NTR4170NT1G

CR0603-16W-000

CR0603-16W-4702F

ON Semiconductor 3.2 A, 30 V Nchan SOT23 Not Installed (NI)

Venkel

0 ,1 A ThickFilm 0603 Not Installed (NI)

47 k, 1/16 W ±1% ThickFilm 0603 Not

Installed (NI)

U4

MC7805ABD2TG

ON Semiconductor

1 A max LDO D2PAK Not Installed (NI)

26

Rev. 0.1

AN726

NOTES:

Rev. 0.1

27

AN726

CONTACT INFORMATION

Silicon Laboratories Inc.

400 West Cesar Chavez

Austin, TX 78701

Please visit the Silicon Labs Technical Support web page:

http://www.silabs.com/support

and register to submit a technical support request.

Patent Notice

Silicon Labs invests in research and development to help our customers differentiate in the market with innovative low-power, small size, analog-

intensive mixed-signal solutions. Silicon Labs' extensive patent portfolio is a testament to our unique approach and world-class engineering team.

The information in this document is believed to be accurate in all respects at the time of publication but is subject to change without notice.

Silicon Laboratories assumes no responsibility for errors and omissions, and disclaims responsibility for any consequences resulting from

the use of information included herein. Additionally, Silicon Laboratories assumes no responsibility for the functioning of undescribed fea-

tures or parameters. Silicon Laboratories reserves the right to make changes without further notice. Silicon Laboratories makes no warran-

ty, representation or guarantee regarding the suitability of its products for any particular purpose, nor does Silicon Laboratories assume any

liability arising out of the application or use of any product or circuit, and specifically disclaims any and all liability, including without limitation

consequential or incidental damages. Silicon Laboratories products are not designed, intended, or authorized for use in applications intend-

ed to support or sustain life, or for any other application in which the failure of the Silicon Laboratories product could create a situation where

personal injury or death may occur. Should Buyer purchase or use Silicon Laboratories products for any such unintended or unauthorized

application, Buyer shall indemnify and hold Silicon Laboratories harmless against all claims and damages.

Silicon Laboratories and Silicon Labs are trademarks of Silicon Laboratories Inc.

Other products or brandnames mentioned herein are trademarks or registered trademarks of their respective holders.

28

Rev. 0.1


型号：	AN726
厂家：	SILICON
描述：	UTILITY CLASS-D TOOLSTICK USERâS GUIDE 实用类-D TOOLSTICK USERA ????指南
文件：	总28页 (文件大小：780K)
中文：	中文翻译
下载：	下载PDF数据表文档文件

AN726 [SILICON]

相关型号：

AN7261FBQ

AN7266

AN727

AN7270

AN7273

AN7273AN7273S

AN7273S

AN7277

AN728

AN7280S

AN7283S

AN7286S