YAMAHA YMU759 Datasheet

YMU759

MA-2

Outline

YMU759 is a synthesis LSI for portable telephone that is capable of playing high quality music by utilizing
FMsynthesizer and ADPCM decorder that are included in this device. As a synthesis, YMU759 is equipped
with Yamaha’s original FM synthesizer, with which the device is capable of simultaneously generating up to
16 voices with different tones. Since the device is capable of generating ADPCM data simultaneously
synchronous with the play of the FM synthesizer, various sampled voices can be used as sound effects.

Since the play data of YMU759 are interpreted at anytime through FIFO, the length of the data (playing period)
is not limited, so the device can flexibly support applications such as incoming call melody distribution service.
The hardware sequencer built in this device allows playing of complex music without giving excessive load to
the CPU of the portable telephones. Moreover, the registers of the FM synthesizer can be operated directly for
real time sound generation, allowing, for example, utilization of various sound effects when using the game
software installed in the portable telephone.

YMU759 includes a speaker amplifier with low ripple whose maximum output is 550 mW
(SPVDD=3.6V). The device is also equipped with conventional functions including a vibrator and a circuit for
controlling LEDs synchronous with music.
For the headphone, it is provided with a stereophonic analog output terminal.

For the purpose of enabling YMU759 to demonstrate its full capabilities, Yamaha proposes to use “SMAF:
Synthetic music Mobile Application Format” as a data distribution format that is compatible with multimedia.
Since the SMAF takes a structure that sets importance on the synchronization between sound and images,
various contents can be written into it including incoming call melody with words that can be used for training
karaoke, and commercial channel that combines texts, images and sounds, and others. The hardware sequencer of
YMU759 directly interprets and plays blocks relevant to systhesis (playing music and reproducing ADPCM with FM

synthesizer) that are included in the data distributed in SMAF.

Features

FM synthesizer functions

Tones

FM synthesizer is capable of creating countless tones theoretically.
When synthesizing tones, it is necessary to designate the number of operators to be used for the synthesis.
(Refer to “5-7. FM synthesis section” for explanation of operator.)
Increasing the number of operators allows synthesis of tones that are more intricate and closer to those

generated by natural musical instruments.
YMU759 supports synthesis of tones of two types including 2-operator tones and 4-operator tones.
Because operator's wave shape can be chosen from eight kinds, the quality of sound improves more
remarkably than 2 operator sound of the MA-1 series. (A MA-1 series can choose operator wave shape from
two kinds.)
Number of voices simultaneously generated
YMU759 is equipped with 32 operators.
The number of voices simultaneously generated varies depending on how many 2-operator tones and
4-operator tones are used.

YAMAHA CORPORATION

YMU759 CATALOG

CATALOG No.:LSI-4MU759A2

2001.1

YMU759

When only 2-operator tones are used: up to 16 voices can be generated simultaneously.
When only 4-operator tones are used: up to 8 voices can be generated simultaneously.
Compatible with stereophonic sound generation.
Volume control
Channel volume, master volume, expression, and pan pot control in individual channels
Sequencer is built in.
Can interpret Mobile Multimedia Format directly.
Equipped with four systems of 96 FIFOs for sequence data
Supports direct access that directly controls FM synthesizer.
Supports key control with half an octave higher and lower.

ADPCM reproduc tion function

Equipped with ADPCM decoder with 4 bits, 1 channel
Supports two kinds of sampling frequency, 4 kHz and 8 kHz.
Sequencer is built in.
Equipped with 348 byte FIFO for ADPCM data and 32 byte FIFO for sequence data
Supports direct access that directly controls ADPCM section .

Speaker amplifier and equalizer circuit

Output of speaker amplifier: 550 mW when SPVDD=3.6 V, or 400 mW when SPVDD=3.0 V
Balanced input speaker amplifier provides low ripple
Built-in equalizer circuit corrects the difference of frequency response among the speakers and forms of
bodies.

Interface

4 wire serial interface or 12 wire parallel in terface can be selected.

Others

PLL is built-in to support master clock input in 2 MHz to 20 MHz range.
Provided with a circuit for controlling on/off of LEDs and vibrator. These can be operated synchronous with
the play data.
Provided with a stereophonic analog output terminal for headphone
16 bit stereophonic D/A converter is built in.
Supports power down mode. (Typical current: 1 µA or less)

Power supply voltage

The power supply includes two power supply sub-systems, analog power supply devoted to speaker amplifier
and power supply for other sections.
The power supply for the speaker amplifier (SPVDD) supplies voltages in the range 2.7 V
and other power supplies (VDD) voltages in the range 2.7 V ~ 3.3 V (Typ 3.0 V).

32-pin plastic QFN.

4.5 V (Typ 3.6 V),

~

-2-

YMU759

Terminal configuration

SPOUT2

EXT2

D7

D6

D5

D4

D3

D2

25 24 23 22 21 20 19 18 17

SPOUT1

VSS

16
15
14
13
12

11

10

VREF

SPVSS
SPVDD
EQ3
EQ2
EQ1
HPOUT-R
HPOUT-L/MONO

D1
D0
/WR
SDIN (/CS)
SYNC (A0)
SCLK (/RD)
SDOUT

26
27
28
29
30
31
32

1 2 3 4 5 6 7 8 9

IFSEL

/RST

/IRQ

EXT1

CLKI

PLLC

VDD

<32pin QFN Top View>

-3-

YMU759

Terminal functions

No. Name I/O Function

1CLKIIsh
2

EXT1

3/IRQO
4/RSTIsh
5 IFSEL I

6 PLLC A

7VDD­8VSS­9VREFA

10 HPOUT-L / MONO A
11 HPOUT-R A
12 EQ1 A
13 EQ2 A
14 EQ3 A

15

SPVDD

16 SPVSS ­17 SPOUT1 A
18 SPOUT2 A
19 EXT2 O
20
21
22

D7
D6
D5

I/O
I/O

I/O
23 D4 I/O
24 D3 I/O
25 D2 I/O
26 D1 I/O
27 D0

I/O Parallel I/F data bus 0 (To be open when IFSEL=L)

28 /WR Ish

29 SDIN (/CS) Ish

30 SYNC (A0) Ish

31 SCLK (/RD) Ish

32 SDOUT OD

Comment: Ish= Schmitt input, OD= open drain terminal, A= Analog terminal
(*) The function changes by setup of the register.

Clock input (2~20MHz)

O

External device control terminal 1 (*)
Interruption output

Hardware reset input
CPU I/F selection L: Serial I/F, H: Parallel I/F

Connection of capacitor for built in PLL

Connect the 3.3kΩ resistance and the 1000pF capacitor between this terminal
and VSS in series.
Digital power supply (Typically +3.0V)
Connect 0.1

Ground

Analog reference voltage.

Connect 0.1
Headphone L channel output: can be switched to mono through register setting
Headphone R channel output
Equalizer terminal 1
Equalizer terminal 2
Equalizer terminal 3
Analog power supply (Typically +3.6 V)

-

Connect 0.1
Analog ground for speaker amplifier
Speaker terminal 1
Speaker terminal 2
External device control terminal 2 (*)
Parallel I/F data bus 7 (*)
Parallel I/F data bus 6 (*)
Parallel I/F data bus 5 (*)
Parallel I/F data bus 4 (To be open when IFSEL=L)
Parallel I/F data bus 3 (To be open when IFSEL=L)
Parallel I/F data bus 2 (To be open when IFSEL=L)
Parallel I/F data bus 1 (To be open when IFSEL=L)

Parallel I/F write puls e (To be open when IFSEL=L)
IFSEL= L Serial I/F data input
IFSEL= H Parallel I/F chip select input
IFSEL= L Serial I/F data decision signal
IFSEL= H Parallel I/F address sign al
IFSEL= L Serial I/F bit clock input
IFSEL= H Parallel I/F read pulse

Serial I/F data output (Pull up resistance is necessary for the outside)

F and 4.7 µF capacitors between this terminal and VSS

µ

F capacitor between this terminal and VSS

µ

F and 4.7 µF capacitors between this terminal and SPVSS

µ

-4-

YMU759

Block diagram

HPOUT-L

/MONO

HPOUT-R

EQ1

EQ

Vol

HP

Vol L

Select

HP

Vol R

Rch

EQ2

-

+

VREF

Lch

EQ3

SP

Vol

SPOUT1

SPOUT2

SPVDD
SPVSS

Analog power

supply devoted

to speaker

amplifier

16 - bit DAC

Lch

Rch

VREF

VREF

CLKI

PLLC

VSS

VDD

Vol

Vol

&

Vol

LPF

FM

(Fs=49.7kHz)

simultaneously

Synthesizer

16 sound gene rated

Timing Generator

Sequencer

Sequencer

ADPCM

Playback

(Fs=4 or 8kHz)

Vibrat o r co n t ro l

LED co n t rol

TIMER

PLL

Power

Down

Control

×4

FM

FIFO

Register

CPU I/F

SELECT

Seq

FIFO

ADPCM

Wave

FIFO

ADPCM

4

Mix & Select

/RST

SCLK SYNC

SDIN

SDOUT

A0

/CS

/WR

/RD

D0 -D7

/IRQ

IFSEL

EXT1 EXT2

-5-

YMU759

Outline of blocks

Explanation about outline of built-in each blocks and flow of the signal are follows.

Clock
Generate

CPU Interface

Register

Hardware
Sequencer

FIFO

FM
Sound
Generator

ADPCM
Playback

DAC

Headphone
Output

EQ
Amplifier

Speaker
Amplifier

CPU interface

Receives commands send from external CPU, int erprets t he co nt ents, and th en writ es them int o regi sters by in d ex
address. Controls reading of designated register data.
As interfaces for controlling YMU759, 4 wire serial and 12 wire parallel interfaces are provided, which can be
selected through IFSEL terminal.

Registers

Register groups that control the LSI except for sequence data.
FM tone register data, various volumes and other control data are store here.

FIFO

Sequence data to move hardware sequencer and ADPCM wave data are stored in FIFO.
This device is equipped with four FIFOs for FM and two FIFOs for ADPCM.
The FIFOs for FM stores sequence data and those for ADPCM stores sequence and waveform data. The size of
FIFOs for FM is 96 bytes, the one for ADPCM data is 384 bytes, and the one for sequence data is 32 bytes.

External
Parts

Hardware sequencer

FIFO is provided as a previous stage of the sequencer which reads sequence data from FIFO to control FM and
ADPCM sections.
The sequence data are compatible with SMAF(Synthetic music Mobile Application Format) proposed by yamaha.

FM synthe sis

This is a synthesis that uses Yamaha’s original FM system. It is able to generate up to 16 voices simultaneously.

This section plays in accordance with commands from the sequencer.

It can also play by directly controlling various registers without using the sequencer.

The sampling frequency is 49.7 kHz that complies with stereophonic sound.

ADPCM playback

This section decodes 4 bit ADPCM data to 16 bit data by using the sampling frequency of 4 kHz or 8 kHz.

It can playback one voice. It playback according to command from sequen cer.

And it can playback to control various register directly without using sequencer.

-6-