YM2413 FM Operator Type-LL (OPLL) Application Manual

Table of Contents

1. What is OPLL?
   1. Outline
   2. Features
   3. Sound Synthesis by FM
2. Functions
   1. Primary Functions
   2. Pin Assignments
   3. Pin Functions
   4. Bus Control
   5. Channels and Slots
   6. Block Diagram
   7. Register Map
3. Description of Operation
   1. Registers
      1. TEST: Address [$0F]
      2. AM/VIB/EG-TYP/KSR/MUL: Address [$00~$01]
      3. KSL/TL/DC/DM/FB: Address [$02~$03]
      4. AR/DR: Address [$04~$05]
      5. SL/RR: Address [$06~$07]
      6. BLOCK/F-Number/SUS/KEY: Address [$A0~$B8]
      7. RHYTHM: Address [$0E]
      8. INST/VOL: Address [$30~$38]
   2. Phase Generator (PG)
   3. Envelope Generator (EG)
   4. Operator (OP) and DAC
4. Interfacing
   1. Clock Generation
   2. Audio Output Interface
   3. Interfacing Microprocessor/Microcomputer
5. Synthesis of Music Sounds
   1. Procedure of Synthesizing Sounds
   2. Fundamentals of Sound Synthesis
   3. Examples of Sound Synthesis
   4. Procedure of Percussion Sound Synthesis
6. Electrical Characteristics
7. Timing Diagrams
8. Outline Dimensions

Original page numbers

• FM Sound Generator for real sound creation.
• Two selectable modes: 9 simultaneous sounds or 6 melody sounds plus 5 rhythm sounds (different tones can be used together in either case).
• Built-in Instruments data (15 melody tones, 5 rhythm tones, "CAPTAIN" and TELETEXT applicable tones).
• Built-in DA Converter.
• Built-in Quartz Oscillator.
• Built-in Vibrato Oscillator/AM Oscillator.
• TTL Compatible Input.
• Si-Gate NMOS LSI.
• A single 5V power source.

This method is expressed with four parameters as follows.

F = A sin (ωct + I sin (ωmt))

(1)

Where A is the amplitude of output, I modulation index, ω_c and ω_m angular frequencies of carrier and modulator respectively. The equation (1) can be expressed alternately as follows.

F = A (J0(I) sin (ωct) + J1(I) (sin ((ωc + ωm) t) - sin ((ωc - ωm) t) + J2(I) (sin ((ωc + 2ωm) t) - sin ((ωc - 2ωm) t) + ...

(2)

Where J_n(I) is the nth-order Bessel function of the first kind. The amplitude of each harmonic component is expressed as a Bessel function of the modulation index. Sound synthesis by FM is found to be useful to obtain specific musical sounds or sound effects. String sounds, however, cannot be obtained as the distribution of harmonics is not uniform. A method called

F = A sin (ωct + βF)

(3)

Where β is the rate of feedback. The spectrum of harmonics produced by "feedback FM" has a saw-tooth waveform and therefore it can be used to synthesize string sounds.

Three function blocks, as follows, are needed to realize the above method of sound synthesis by FM.

1. Phase generator (PG) generating ω_c
2. Envelope generator (EG) generating amplitude A and modulation index (I) which vary with time
3. Sin table (SIN)

The method of sound synthesis by FM can be realized by Figure I-1 with units (operator cells) which combine the three function blocks. By this method, all we have to do is to define a frequency parameter and EG parameters.

Figure I-1 Sound synthesis by FM with units

II. Functions

FM = E sin (ω₁t + I sin (ω₂t))

Table II-1 Mode selection

	CS	WE	A0
1	1	*	*	Inactive mode
2	0	0	0	Address write mode
3	0	0	1	Data write mode
4	0	1	0	Bus inhibited

1. Inactive mode
   The bus lines (D₀~D₇) are at high impedance while CS is '1'.
2. Address Write mode
   When writing an address, set the control signals to the address write mode and place the address data on the data bus. The address of the specified internal register is now ready to receive data. After writing an address, be sure to wait 12 cycles of the master clock (øM) before writing in sound data.
3. Data write mode
   When the control signals as (are) set for the data write mode, the data existing on the data bus enters the register of the given address. Wait for 84 cycles of the master clock (øM) before writing in further data or another address.
4. Bus inhibited
   In this mode, the data of (on) the data bus is meaningless and uncontrollable.

OPLL needs a certain length of wait time between successive writings of addresses or data into the registers. The wait time varies between address and data write modes. The processor waits for a time in Table II-2 before OPLL performs another function. If the wait time is not observed data will be uncertain.

Table II-2 Wait times

Mode	Wait time
Address write	12 cycles
Data write	84 cycles

Note: the wait times are given in cycles of the master clock.

Channel data, like the F-number, controls two slots, No. 1 and 2.

In FM mode, the No. 1 slot handles the modulating waves and the No. 2 slot the carrier. The No. 1 slot can also be set for feedback FM mode. See (III-1-3) for setting this mode.

Table II-3 Channels and slots

1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	Slot No.
1	2	3	1	2	3	4	5	6	4	5	6	7	8	9	7	8	9	Channel No.
1			2			1			2			1			2			Slot No. when a channel is seen
20	21	22	20	21	22	23	24	25	23	24	25	26	27	28	26	27	28	Relationship between the data of each channel and the register. (Example $20 to $28)

(II-6) Block Diagram

Register map for Rhythm mode (addr. $0E D₅ = 'H')

Address	D7	D6	D5	D4	D3	D2	D1	D0
36					BD-VOL				Rhythm volume register
37	HH-VOL				SD-VOL
38	TOM-VOL				T-CY-VOL

	Address	Bit
1	00 (m) 01 (c)	D7	Amplitude modulation ON/OFF switch
		D6	Vibrato ON/OFF switch
		D5	Sustained tone/decaying tone switch. 0: Percussive tone 1: sustained tone
		D4	RATE key scale
		D0~D3	Controls Multiple sample wave-harmonics relationship
2	02 (m) 03 (c)	D6 D7	LEVEL key scale
3	02	D0~D5	Modulator total level. Modulation index control
4	03	D4 (c) D5 (m)	Carrier and modulated wave distortion waveform (flat wave rectification) ON/OFF switch
		D0~D2	FM feedback constant (m)
5	04 (m) 05 (c)	D4~D7	Attack envelope change rate control
		D0~D3	Decay envelope change rate control
6	06 (m) 07 (c)	D4~D7	Indication of decay-sustain level
		D0~D3	Release envelope change rate control
7	0E	D5	Rhythm sound mode selection. 1: Rhythm sound mode 0: Melody sound mode
		D0~D4	Rhythm instruments ON/OFF switch
8	10~18	D0~D7	F-Number LSB 8 bits
9	20~28	D5	Sustain ON/OFF switch
		D4	Key ON/OFF switch
		D1~D3	Octave setting
		D0	F-Number MSB
10	30~38	D4~D7	Instruments selection
		D0~D3	Volume data

Tone Data

INST	Instrument
0	Original
1	Violin
2	Guitar
3	Piano
4	Flute
5	Clarinet
6	Oboe
7	Trumpet
8	Organ
9	Horn
10	Synthesizer
11	Harpsichord
12	Vibraphone
13	Synthesizer Bass
14	Acoustic Bass
15	Electric Guitar

BD	Bass Drum
SD	Snare Drum
TOM	Tom-tom
T-CY	Top Cymbal
HH	High Hat

All registers are cleared to 0 by initialization (initial clear IC terminal = 0).

(III-1-1) TEST: Address [$0F]

Normally all addresses are 0.

(III-1-2) AM/VIB/EG-TYP/KSR/MUL: Address [$00~$01]

D7	D6	D5	D4	D3	D2	D1	D0
AM	VIB	EG-TYP	KSR	MUL
				23	22	21	20

• D₀~D₃ (MUL): The frequencies of the carrier wave and the modulated wave are controlled according to the multiplying factors in Table III-1.

  <Example>

  F(t) = E sin (ω_ft + I sin (7 ω_ft))

  • Frequency according to F-number = ω_ft
    
  • Carrier MUL = 1
    
  • Modulator MUL = 7
  Page 9 - original - top
  Table III-1 Multiplying factors

  MUL	Multiplying factor
  0	½
  1	1
  2	2
  3	3
  4	4
  5	5
  6	6
  7	7
  8	8
  9	9
  A	10
  B	10
  C	12
  D	12
  E	15
  F	15

• D₄ (KSR): This bit specifies the key scale of RATE.

  Since almost all musical steps become high for natural instruments, the sound can quickly change to high or low. This is accomplished by the key scale of RATE. Values in Table III-2 include the speed offset for the musical steps. Therefore, the actual RATE set for the ADSR will be these values which include the offset.

  RATE = 4 × R + R_ks

  • R is the set value for ADSR
  • R_ks is the key scale offset value
  • Note that when R=0, RATE=0

  Table III-2

  0		1		2		3		4		5		6		7		Octave
  0		1		2		3		4		5		6		7		BLOCK Data
  0	1	0	1	0	1	0	1	0	1	0	1	0	1	0	1	F-Num MSB
  0	0	0	0	1	1	1	1	2	2	2	2	3	3	3	3	D4 = 0 key scale
  0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	D4 = 1 key scale

• D₅ (EG-TYP): Switching between sustained tone and Percussive tone.

  When D₅ = '0' it will be percussive tone. When D₅ = '1' it will be sustained tone. These sound modes are different in that the RELEASE RATE usage differs. The difference is shown in Figure III-1.

  Page 10 - original - top
  Percussive tone (D₅ = 0)

  

  Sustained tone (D₅ = 1)

  

  Fig. III-1

• D₄ (VIB): Vibrato on/off switch. When this bit is '1', vibrato will be applied to the slot. The frequency is 6.4 Hz (@ øM = 3.6 MHz).

• D₇ (AM): Amplitude modulation on/off switch. When this bit is '1', amplitude modulation wil be epplied to the slot. The frequency is 3.7 Hz (@ øM = 3.6 MHz).

Total Level is used to control the modulation index (tone) by adding attenuation to the output of the envelope generator. Level key scale (KSL) ensures, like the key scale of RATE, that the output level of natural instruments falls as the pitch of sound rises.

	D7	D6	D5	D4	D3	D2	D1	D0
$02	KSL		TL
$03				DC	DM	FB

• D₀~D₅ (TL): The minimum resolution of the attenuation level is 0.75 dB and the sound level can be reduced down to 47.25 dB.
  Page 11 - original - top
  Table III-3 Total level

  	D5	D4	D3	D2	D1	D0
  Attenuation level	24 dB	12 dB	6 dB	3 dB	1.5 dB	0.75 dB

• D₆~D₇ (KSL): This bit controls the level of key scaling. In key scale mode, the level of attenuation increases as pitch rises, to 1.5 dB/oct, 3 dB/oct, 6 dB/oct, and 0 dB/oct.

  Table III-4

  D7	D6	Attenuation rate
  0	0	0 dB/oct
  1	0	1.5 dB/oct
  0	1	3 dB/oct
  1	1	6 dB/oct

  Table III-5 Attenuation at each F-Number at 3 dB/OCT

  OCT	F-Number
  	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
  0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0	0
  1	0	0	0	0	0	0	0	0	0	0.75	1.125	1.5	1.875	2.25	2.525	3
  2	0	0	0	0	0	1.125	1.875	2.525	3	3.75	4.125	4.5	4.875	5.25	5.625	6
  3	0	0	0	1.875	3	4.125	4.875	5.625	6	6.75	7.125	7.5	7.875	8.25	8.625	9
  4	0	0	3	4.875	6	7.125	7.875	8.625	9	9.75	10.125	10.5	10.875	11.25	11.625	12
  5	0	3	6	7.875	9	10.125	10.875	11.625	12	12.75	13.125	13.5	13.875	14.25	14.825	15
  6	0	6	9	10.875	12	13.125	13.875	14.825	15	15.75	16.125	16.5	16.875	17.25	17.625	18
  7	0	9	12	13.875	15	16.125	16.875	17.625	18	18.75	19.125	19.5	19.875	20.25	20.625	21

  Unit: dB

  Notes:

  • F-Number is the value of the four MSBs.
  • Half of the above data at 1.5 dB/oct
  • Double of the above at 6 dB/oct
  Page 12 - original - top
• $03: D₃ (DM): The modulated wave is rectified to half wave.

• $03: D₄ (DC): The carrier wave is rectified to half wave.

• $03: D₀~D₂ (FB): For the modulated wave of the first slot's feedback FM modulation.

  Table III-6 Modulation index

  	0	1	2	3	4	5	6	7
  Modulation index	0	Π/16	Π/8	Π/4	Π/2	Π	2Π	4Π

  [Π = pi in this font]

	D7	D6	D5	D4	D3	D2	D1	D0
	AR				DR
$04	23	22	21	20	23	22	21	20	Modulator
$05	23	22	21	20	23	22	21	20	Carrier

(III-1-5) SL/RR: Address [$06~$07]

Sustain Level is the critical point that, when sustaining sound has been attenuated to the prescribed level, the level is maintained thereafter. With percussive tone, it is the critical point of change from decay mode to release mode.

Release Rate is the disappearing rate of sustaining sounds after key OFF. For percussive tone, attenuation above the sustain level is expressed with decay rate and attenuation below the sustain level with release rate.

	D7	D6	D5	D4	D3	D2	D1	D0
	SL				RR
$06	24 dB	12 dB	6 dB	3 dB	23	22	21	20	Modulator
$07	24 dB	12 dB	6 dB	3 dB	23	22	21	20	Carrier

* Attenuation times of the release rate are the same as that of the decay rate.

Likely transcription errors here, especially lower in the table

Note: When RATE is '0', the envelope will not change.

Data defining a musical scale and interval. The F-Number is contained in the registers $1* and $2*.

	D7	D6	D5	D4	D3	D2	D1	D0
$10~$18	F-Number
	27	26	25	24	23	22	21	20
$20~$28			SUS	KEY	BLOCK			F-Num
					22	21	20	28

• D₀~D₇ [$1*], D₀ [$2*] (F-Number): The F-Number is defined with 8 bits of the $1* register and LSB of $2*, for a total of 9 bits. This F-Number defines a scale (how a value is assigned to the F-number is described later).

• D₁~D₃ (BLOCK): Defines octave information.

• D₄ (KEY): This bit indicates the ON/OFF state of the key. When this bit is '1', the associated channel is on and sound is produced. '0' is equivalent to Key-OFF.

• D₅ (SUS): When this bit is '1', the RR with the key off will be 5.

F-Number/Block

The OPLL generates the necessary frequency when an appropriate increment of phase is given. The increment is determined by the F-Number, Block, and Multiple information. The following formula gives the increment for a desired frequency.

ΔP = fmus × 218 / fsam

(1)

• f_sam = f_M/72
• f_mus: Desired frequency
• f_sam: Sampling frequency (50 kHz)
• f_M: Input clock frequency (3.6 MHz)

ΔP = 2B × F' × MUL

(2)

• B: Octave information
• F': Increment within an octave
• MUL: Multiple data

F = (f_mus × 2¹⁸/f_sam) / 2^b-1       @ MUL = 1

• F: F-Number data
• b: Block data

Table III-8-1 F-Number (1)

Sound steps	Frequency (4 oct)	F-Number	$2*	$1*
			D0	D7	D6	D5	D4	D3	D2	D1	D0
C#	277.2	181	0	1	0	1	1	0	1	0	1
D	293.7	192	0	1	1	0	0	0	0	0	0
D#	311.1	204	0	1	1	0	0	1	1	0	0
E	329.6	216	0	1	1	0	1	1	0	0	0
F	349.2	229	0	1	1	1	0	0	1	0	1
F#	370.0	242	0	1	1	1	1	0	0	1	0
G	392.0	257	1	0	0	0	0	0	0	0	1
G#	415.3	272	1	0	0	0	1	0	0	0	0
A	440.0	288	1	0	0	1	0	0	0	0	0
A#	466.2	305	1	0	0	1	1	0	0	0	1
B	193.9	323	1	0	1	0	0	0	0	0	1
C	523.3	343	1	0	1	0	1	0	1	1	1

To control ON/OFF of the Rhythm mode selection and percussion instruments.

D5	D4	D3	D2	D1	D0
RHYTHM	BD	SD	TOM	TOP-CY	HH

D₀~D₅ (RHYTHM): When D₅ = 1, OPLL is in Rhythm mode with percussion sounds produced through channels 7~9 (see page 5). In this mode, the melody section is limited to six sounds. D₀~D₄ controls ON/OFF of percussion instruments. Therefore Key-ON bits $26, $27, $28 must always be cleared to 0. Slots 13~18 are related with percussion sounds as shown in Table III-9, and data of F-Num must input values that match percussion sounds.

Table III-9 Rhythm slots

Instrument	Slot
BD	13, 16
SD	17
TOM	15
TOP-CYM	18
HH	14

Address	Data
16	20
17	50
18	C0
26	05
27	05
28	01

Determine the voice (15 voices preset in ROM and original voices) and the volume.

	D7	D6	D5	D4	D3	D2	D1	D0
$30~$38	INST				VOL

• D₄~D₇ (INST): These four bits determine voices as follows:

  INST	Instrument
  0	Original
  1	Violin
  2	Guitar
  3	Piano
  4	Flute
  5	Clarinet
  6	Oboe
  7	Trumpet
  8	Organ
  9	Horn
  A	Synthesizer
  B	Harpsichord
  C	Vibraphone
  D	Synthesizer Bass
  E	Acoustic Bass
  F	Electric Guitar

• D₀~D₃ (VOL): Determine the volume for the voices. The minimum resolution is 3 dB and the maximum 45 dB.

  D3	D2	D1	D0
  24 dB	12 dB	6 dB	3 dB

  In the rhythm mode (addr = $0E, D₅ = 'H'), each rhythm volume for $36~$38 are set as follows:

  Addr	D7	D6	D5	D4	D3	D2	D1	D0
  $36					BD
  $37	HH				SD
  $38	TOM				T-CYM

Figure III-2 Envelope waveform

The DAC performs DA conversion for all the sounds then outputs them as shown in Figure III-3 (a). To sum up the sounds, an integrated circuit is added to MO/RO. (Figure III-3 (b)) Since the level of percussive sounds seems lower, when compared to musical sounds, the same percussive sounds are output twice. (Figure III-3 (c))

Figure III-3

Figure IV-1 System block diagram

(IV-1) Clock Generation

Figure IV-3 Interfacing processor

Since the FM's basic parameter (carrier output level, modulator output level, modulator feedback level, carrier frequency, modulator frequency) is handled effectively, the pitch, tone and volume of sounds can all be decided. Table V-1 shows the relationship between FM parameters and OPLL parameters.

(V-3) Examples of Sound Synthesis

1. Electric piano
   1. Determination of operator frequencies

      Set MUL to '1', for both operators, to produce an overtone of higher frequency. (Increases are integer multiples)

      Page 23 - original - top
   2. Output level of the operator

      Change the output of the modulator to adjust the tone. When determining the level of operator 1, adjust the sounds for rich overtones like those of the piano. Then make adjustment for sounds of higher frequencies by level scaling of operator 1. Level scaling is necessary so that the waveforms are nearly sine waves in high frequency ranges.

   3. Setting EG

      Determine sound volume, tone, and envelope. Set operator 2 so that the sounds have a sharp attack and long sustain. (This can be set varying degrees.) Let operator 1, the modulator, produce many overtones at the attack time and later keep a certain composition of overtones unchanged. To adjust sound volume, apply key scaling to operator 2. Perform RATE scaling to make sounds of higher pitches sharp.

   4. Readjustment of data

      Now sound synthesis is almost completed. Tones will vary somewhat depending on the setting of EG. Make a final adjustment with the output level and feedback level of the operators. If the sound is too metallic, for example, resuce the level of operator 1.

   5. Adding effect

      Finally, add tremolo effect by LFO to simulate the sounds of the electric piano. This is possible by using the built-in amplitude modulation function or by refreshing the value of TOTAL LEVEL in intervals of 2~6Hz (possibly with triangular waves) by software.

2. Trumpet
   1. Output of operator

      Adjust the total level of operator 1, the modulator, at $10~$28 and the feedback level to the maximum of '7' for bright tones.

   2. Operator frequencies

      Set both operators at '1'.

   3. EG

      Adjust two EGs for a slow attack. For brass sounds, ensure the modulator attack is slower than the carrier's. This is necessary to express the characteristic attack of brass sounds.

   4. Key scaling

      High-pitched sounds are not crisp because the envelope is set for a slow rise. Apply rate scaling for a more natural feeling of fast passages.

      Page 24 - original - top
   5. LFO

      With a brass instrument, the pitch of long notes fluctuates a little even when a good player is playing. To express this, apply vibrato effect.

OPLL includes a noise oscillator obtained by composing a white noise generator with several frequencies. This noise oscillator is specified by the frequency information (BLOCK, F-Number, Multiple) of the 8 and 9 channels. When composed with white noise, phase output suitable for percussion instruments is generated and given to the operator. Thus phases of four instruments are generated from two sets of frequency information. It is known empirically that the optimum ration of two frequencies is 3:1 (f_7CH = 3×f_8CH). Now phase data of individual instruments are obtained. Next, we multiply this output with envelope information. Since the envelope is set as 1 slot -> 1 percussion instrument, values that reflect the characteristics of individual percussion instruments, as with melody instruments, are set in ROM. (Refer to III-1-7)

1. Absolute Maximum Ratings

   ITEM	RATING	UNIT
   Pin voltage	0.3~7.0	V
   Ambient operating temperature	0~70	°C
   Storage temperature	-50~125	°C

2. Recommended Operating Conditions

   ITEM	SYMBOL	MIN.	TYP.	MAX.	UNIT
   Supply voltage	VCC	4.75	5	5.25	V
   	GND	0	0	0	V

3. DC Characteristics

   ITEM		SYMBOL	CONDITION	MIN.	TYP.	MAX.	UNIT
   High level input voltage	All input	VIH		2.0		VCC	V
   Low level input voltage	All input	VIL		-0.3		0.8	V
   Leak input current	A0, WE	IL	Vin = 0~5 V	-10		10	µA
   Three-state (off) input current	D0~D7	ITSL	Vin = 0~5 V	-10		10	µA
   Analog output voltage	MO	VMOA	RLOAD = 2.2K		1.6		VPP
   	RO	VROA	RLOAD = 2.2K		1.6		VPP
   Pullup resistance	IC, CS	RPU		100			KΩ
   Input capacity	All input	CI				10	pF
   Output capacity	All input	CO				10	PF
   Power current		ICC			40	80	MA

4. AC Characteristics

   ITEM		SYMBOL	CONDITION	MIN.	TYP.	MAX.	UNIT
   Address setup time	A0	TAS	Fig. A-1	10			ns
   Address hold time	A0	TAH	Fig. A-1	10			ns
   Chip select write width	CS	TCSW	Fig. A-1	80			ns
   Write pulse write width	WE	TWW	Fig. A-1	110			ns
   Write pulse setup time	WE	TWS	Fig. A-1	30			ns
   Write data setup time	D0~D7	TWDS	Fig. A-1	10			ns
   Write data hold time	D0~D7	TWDH	Fig. A-1	25			ns
   Reset pulse width	IC	NICW	Fig. A-2		80		Cycle

   Page 26 - original - top
5. DAC properties

   ITEM		SYMBOL	CONDITION	MIN.	TYP.	MAX.	UNIT
   Maximum output oscillation	RO, MO	Vout	Fig. IV-2		2/5 VCC		VPP
   Resolution	RO, MO		Fig. IV-2		9		Bit
   Noise	RO, MO		Fig. IV-2		-65		dB

   Note: The noise level should be suitable for the volume level.

NOTE: T_CSW, T_WW and T_WDH have been measured with either CS or WE high.
Fig. A-1 Write Timing

Fig. A-2 Reset Timing