a Voiceband Signal Port
AD28msp02
Complete Analog I/O Port for Voiceband DSP Applications
Linear-Coded 16-Bit Sigma-Delta ADC Linear-Coded 16-Bit Sigma-Delta DAC On-Chip Anti-Aliasing and Anti-lmaging Filters On-Chip Voltage Reference
8 kHz Sampling Frequency
Twos Complement Coding 65 dB SNR + THD
Programmable Gain on DAC and ADC Serial Interface To DSP Processors 24-Pin DlP/28-Lead SOIC
Single 5 V Power Supply
FUNCTIONAL BLOCK DIAGRAM |
||||
VOICEBAND |
|
|
|
|
ANALOG |
|
|
|
|
INPUT A |
|
|
16-BIT |
|
|
|
+20dB |
||
MUX |
SIGMA- |
|||
AMP |
||||
|
|
DELTA ADC |
||
|
|
|
||
VOICEBAND |
|
|
|
|
ANALOG |
|
|
DIGITAL |
|
INPUT B |
|
|
DATA AND |
|
|
|
|
CONTROL |
|
|
VOLTAGE |
SERIAL |
||
|
REFERENCE |
PORT |
||
DIFFERENTIAL |
|
|
16-BIT |
|
ANALOG |
|
PGA |
SIGMA- |
|
OUTPUT |
|
|
DELTA DAC |
GENERAL DESCRIPTION
The AD28msp02 Voiceband Signal Port is a complete analog front end for high performance voiceband DSP applications. Compared to traditional μ-law and A-law codecs, the AD28msp02’s linear-coded ADC and DAC maintain wide dynamic range while maintaining superior SNR and THD. A sampling rate of 8.0 kHz coupled with 65 dB SNR + THD performance make the AD28msp02 attractive in many telecom and speech processing applications, for example digital cellular radio and high quality telephones. The AD28msp02 simplifies overall system design by requiring only a single +5 V power supply.
The inclusion of on-chip anti-aliasing and anti-imaging filters, 16-bit sigma-delta ADC and DAC, and programmable gain amplifiers ensures a highly integrated and compact solution to voiceband analog processing requirements. Sigma-delta conversion technology eliminates the need for complex off-chip antialiasing filters and sample-and-hold circuitry.
The AD28msp02’s serial I/O port provides an easy interface to host DSP microprocessors such as the ADSP-2101, ADSP-2105 and ADSP-2111. The AD28msp02 is available in a 24-pin, 0.3" plastic DIP and a 28-lead SOIC package.
Figure 1 shows a block diagram of the AD28msp02.
The A/D conversion circuitry of the AD28msp02 consists of two analog input amplifiers, an optional 20 dB preamplifier, and a sigma-delta analog-to-digital converter (ADC). The analog input signal to the AD28msp02 must be ac-coupled.
REV. 0
Information furnished by Analog Devices is believed to be accurate and reliable. However, no responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other rights of third parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices.
Analog Input Amplifiers
The two analog input amplifiers (NORM, AUX) are internally biased by an on-chip voltage reference in order to allow operation of the AD28msp02 with a single +5 V power supply.
An analog multiplexer selects either the NORM or AUX amplifier as the input to the ADC’s sigma-delta modulator. The optional 20 dB preamplifier may be used to increase the signal level; the preamplifier can be inserted before the modulator or can be bypassed. Input signal level to the sigma-delta modulator
should not exceed VINMAX, which is specified under “Analog Interface Electrical Characteristics.” Refer to “Analog Input” in
the “Design Considerations” section of this data sheet for more information.
The input multiplexer and 20 dB preamplifier are configured by Bits 0 and 1 (IPS, IMS) of the AD28msp02’s control register. If the multiplexer setting is changed while an input signal is being processed, the ADC’s output must be allowed time to settle to ensure that the output data is valid.
ADC
The ADC consists of a 2nd-order analog sigma-delta modulator, an anti-aliasing decimation filter, and a digital high-pass filter. The sigma-delta modulator noise-shapes the signal and produces 1-bit samples at a 1.0 MHz rate. This bit stream, which represents the analog input signal, is fed to the anti-aliasing decimation filter.
Decimation Filter
The anti-aliasing decimation filter contains two stages. The first stage is a sinc4 digital filter that increases resolution to 16 bits and reduces the sample rate to 40 kHz. The second stage is an IIR low-pass filter.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A. Tel: 617/329-4700 Fax: 617/326-8703
AD28msp02 |
|
|
|
|
|
|
|
VFB NORM |
|
|
|
|
|
|
|
VINNORM |
|
16-BIT SIGMA-DELTA ADC |
|
|
SDOFS |
||
NORM |
|
|
|
||||
|
|
|
|
|
|
|
|
INPUT |
|
ANALOG |
1 |
ANTI-ALIASING |
16 |
DIGITAL |
16 |
AMP |
+20dB |
||||||
MUX |
SIGMA-DELTA |
|
DECIMATION |
|
HIGH-PASS |
SDO |
|
AMP |
|
|
|||||
VFB AUX |
MODULATOR |
|
FILTER |
|
FILTER |
||
|
|
1.0 |
8.0 |
8.0 |
|||
VINAUX |
|
|
|
|
|||
|
|
MHz |
|
kHz |
|
kHz |
|
AUX |
|
|
|
|
|
|
|
INPUT |
|
|
|
|
|
|
DATA/ |
AMP |
|
|
|
|
|
|
CNTRL |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
CONTROL |
|
|
SERIAL |
|||||||
VREF |
VOLTAGE |
|
|
|
|
|
|
|
|
|
|
|
|
|
REGISTER |
|
|
PORT |
|||||||||||
REFERENCE |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
SCLK |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
||
|
|
|
|
|
|
|
|
|
|
|
|
|
16-BIT SIGMA-DELTA DAC |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
VOUTP |
|
|
|
|
|
|
|
|
|
ANALOG |
1 |
DIGITAL |
16 |
ANTI-IMAGING |
16 |
|
|
DIGITAL |
|
16 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
||||||||||
|
|
|
|
|
|
|
|
PGA |
|
|
SMOOTHING |
|
SIGMA-DELTA |
|
INTERPOLATION |
|
|
|
HIGH-PASS |
|
|
|
|
|
|
|
|
|
SDI |
VOUTN |
|
|
|
|
|
|
|
|
|
FILTER |
1.0 |
MODULATOR |
1.0 |
FILTER |
8.0 |
|
|
FILTER |
|
8.0 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
||||||||||
|
|
|
OUTPUT |
|
|
|
|
|
MHz |
|
MHz |
|
kHz |
|
|
|
|
kHz |
|
|
|
|
|
|
|
SDIFS |
|||
|
|
|
DIFFERENTIAL |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|||||
|
|
|
AMP |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
CS |
|||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Figure 1. AD28msp02 Block Diagram
The IIR low-pass filter is a 10th-order elliptic filter with a passband edge at 3.7 kHz and a stopband attenuation of 65 dB at 4 kHz. This filter has the following specifications:
The high-pass filter is a 4th-order elliptic filter with a passband cutoff at 150 Hz. Stopband attenuation is 25 dB. This filter has the following specifications:
Filter type: |
10th-order low-pass elliptic IIR |
Filter type: |
4th-order high-pass elliptic IIR |
Sample frequency: |
40.0 kHz |
Sample frequency: |
8.0 kHz |
Passband cutoff:* |
3.70 kHz |
Passband cutoff: |
150.0 Hz |
Passband ripple: |
±0.2 dB |
Passband ripple: |
±0.2 dB |
Stopband cutoff: |
4.0 kHz |
Stopband cutoff: |
100.0 Hz |
Stopband ripple: |
–65.00 dB |
Stopband ripple: |
–25.00 dB |
*The passband cutoff frequency is defined to be the last point in the passband that meets the passband ripple specification.
(Note that these specifications apply only to this filter, and not to the entire ADC. The specifications can be used to perform further analysis of the exact characteristics of the filter, for example using a digital filter design software package.)
Figure 2 shows the frequency response of the IIR low-pass filter.
0
–20
dB– |
–60 |
|
|
|
|
|
MAGNITUDE |
|
|
|
|
|
|
|
–40 |
|
|
|
|
|
LOG |
|
|
|
|
|
|
|
–80 |
|
|
|
|
|
–100 |
2600 |
3200 |
3800 |
4400 |
5000 |
|
|
2000 |
FREQUENCY – Hz
Figure 2. IIR Low-Pass Filter Frequency Response
High-Pass Filter
The digital high-pass filter removes frequency components at the low end of the spectrum; it attenuates signal energy below the passband of the converter. The high-pass filter can be bypassed by setting the ADBY bit (Bit 3) of the AD28msp02’s control register.
(Note that these specifications apply only to this filter, and not to the entire ADC. The specifications can be used to perform further analysis of the exact characteristics of the filter, for example using a digital filter design software package.)
Figure 3 shows the frequency response of the high-pass filter.
0
–20
dB– |
–60 |
|
|
|
|
|
MAGNITUDE |
|
|
|
|
|
|
|
–40 |
|
|
|
|
|
LOG |
|
|
|
|
|
|
|
–80 |
|
|
|
|
|
–100 |
60 |
120 |
180 |
240 |
300 |
|
|
0 |
FREQUENCY – Hz
Figure 3. High-Pass Filter Frequency Response
Passband ripple is ±0.2 dB for the combined effects of the ADC’s digital filters (i.e., high-pass filter and IIR low-pass of the decimation filter) in the 300 Hz–3400 Hz passband.
The output of the ADC is transferred to the AD28msp02’s serial port (SPORT) at an 8 kHz rate, for transmission to the host DSP processor. Maximum group delay in the ADC will not exceed 1 ms in the region from 300 Hz to 3 kHz.
–2– |
REV. 0 |
AD28msp02
Pin Name |
I/O/Z Function |
||||||
|
|
|
|
|
|
|
|
VINNORM |
I |
Analog input to inverting terminal of |
|||||
|
|
|
|
|
NORM input amplifier. |
||
VFBNORM |
O |
Output terminal of NORM amplifier. |
|||||
VINAUX |
I |
Analog input to inverting terminal of |
|||||
|
|
|
|
|
AUX input amplifier. |
||
VFBAUX |
O |
Output terminal of AUX amplifier. |
|||||
VOUTP |
O |
Analog output from noninverting |
|||||
|
|
|
|
|
terminal of differential output amplifier. |
||
VOUTN |
O |
Analog output from inverting terminal of |
|||||
|
|
|
|
|
differential output amplifier. |
||
VREF |
O |
On-chip bandgap voltage reference |
|||||
|
|
|
|
|
(2.5 V ± 10%). |
||
MCLK |
I |
Master clock input; frequency must |
|||||
|
|
|
|
|
equal 13.0 MHz to guarantee listed |
||
|
|
|
|
|
specifications. |
||
SCLK |
O/Z |
Serial clock used to clock data or control |
|||||
|
|
|
|
|
bits to and from the serial port |
||
|
|
|
|
|
(SPORT). The frequency of SCLK is |
||
|
|
|
|
|
equal to the frequency of the master |
||
|
|
|
|
|
clock (MCLK) divided by 5. SCLK is |
||
|
|
|
|
|
3-stated when CS is low. |
||
SDI |
I |
Serial data input of SPORT. Both data |
|||||
|
|
|
|
|
and control information are input on |
||
|
|
|
|
|
this pin. Input at SDI is ignored when |
||
|
|
|
|
|
CS is low. |
||
SDO |
O/Z |
Serial data output of SPORT. Both data |
|||||
|
|
|
|
|
and control information are output on |
||
|
|
|
|
|
this pin. SDO is 3-stated when CS is |
||
|
|
|
|
|
low. |
||
SDIFS |
I |
Framing signal for SDI serial transfers. |
|||||
|
|
|
|
|
Input at SDIFS is ignored when CS is |
||
|
|
|
|
|
low. |
||
SDOFS |
O/Z |
Framing signal for SDO serial transfers. |
|||||
|
|
|
|
|
SDOFS is 3-stated when CS is low. |
||
DATA/ |
|
|
|
I |
Configures AD28msp02 for either data |
||
CNTRL |
|||||||
|
|
|
|
|
or control information transfers (via |
||
|
|
|
|
|
SPORT). |
||
CS |
I |
Active-high chip select. Can be used to |
|||||
|
|
|
|
|
3-state the SPORT interface; when CS |
||
|
|
|
|
|
is low, the SCLK, SDO, and SDOFS |
||
|
|
|
|
|
outputs are 3-stated and the SDI and |
||
|
|
|
|
|
SDIFS inputs are ignored. If CS is de- |
||
|
|
|
|
|
asserted during a serial data transfer, the |
||
|
|
|
|
|
16-bit word being transmitted is lost. |
||
|
|
I |
Active low reset signal; resets Control |
||||
RESET |
|||||||
|
|
|
|
|
Register and clears digital filters. |
RESET |
|
|
|
|
|
|
does not 3-state the SPORT outputs |
||
|
|
|
|
|
(SCLK, SDO, SDOFS). |
||
VCC |
|
Analog supply voltage; nominal +5 V. |
|||||
GNDA |
|
Analog ground. |
|||||
VDD |
|
Digital supply voltage; nominal +5 V. |
|||||
GNDD |
|
Digital ground. |
The D/A conversion circuitry of the AD28msp02 consists of a sigma-delta digital-to-analog converter (DAC), an analog smoothing filter, a programmable gain amplifier, and a differential output amplifier.
DAC
The AD28msp02’s sigma-delta DAC implements digital filters and a sigma-delta modulator with the same characteristics as the filters and modulator of the ADC. The DAC consists of a digital high-pass filter, an anti-imaging interpolation filter, and a digital sigma-delta modulator.
The DAC receives 16-bit samples from the host DSP processor via AD28msp02’s serial port at an 8 kHz rate. If the host processor fails to write a new value to the serial port, the existing (previous) data is read again. The data stream is filtered first by the DAC’s high-pass filter and then by the anti-imaging interpolation filter. These filters have the same characteristics as the ADC’s anti-aliasing decimation filter and digital high-pass filter.
The output of the interpolation filter is fed to the DAC’s digital sigma-delta modulator, which converts the 16-bit data to 1-bit samples at a 1.0 MHz rate. The modulator noise-shapes the signal such that errors inherent to the process are minimized in the passband of the converter. The bit stream output of the sigmadelta modulator is fed to the AD28msp02’s analog smoothing filter where it is converted to an analog voltage.
High-Pass Filter
The digital high-pass filter of the AD28msp02’s DAC has the same characteristics as the high-pass filter of the ADC. The high-pass filter removes frequency components at the low end of the spectrum; it attenuates signal energy below the passband of the converter. The DAC’s high-pass filter can be bypassed by setting the DABY bit (Bit 2) of the AD28msp02’s control register.
The high-pass filter is a 4th-order elliptic filter with a passband cutoff at 150 Hz. Stopband attenuation is 25 dB. This filter has the following specifications:
Filter type: |
4th-order high-pass elliptic IIR |
Sample frequency: |
8.0 kHz |
Passband cutoff: |
150.0 Hz |
Passband ripple: |
±0.2 dB |
Stopband cutoff: |
100.0 Hz |
Stopband ripple: |
–25.00 dB |
(Note that these specifications apply only to this filter, and not to the entire DAC. The specifications can be used to perform further analysis of the exact characteristics of the filter, for example using a digital filter design software package.)
Figure 3 shows the frequency response of the high-pass filter.
Interpolation Filter
The anti-imaging interpolation filter contains two stages. The first stage is an IIR low-pass filter that interpolates the data rate from 8 kHz to 40 kHz and removes images produced by the interpolation process. The output of this stage is then interpolated to 1.0 MHz and fed to the second stage, a sinc4 digital filter that attenuates images produced by the 40 kHz to 1.0 MHz interpolation process.
REV. 0 |
–3– |
AD28msp02
The IIR low-pass filter is a 10th-order elliptic filter with a passband edge at 3.70 kHz and a stopband attenuation of 65 dB at 4 kHz. This filter has the following specifications:
Filter type: |
l0th-order low-pass elliptic IIR |
Sample frequency: |
40.0 kHz |
Passband cutoff:* |
3.70 kHz |
Passband ripple: |
±0.2 dB |
Stopband cutoff: |
4.0 kHz |
Stopband ripple: |
–65.00 dB |
*The passband cutoff frequency is defined to be the last point in the passband that meets the passband ripple specification.
(Note that these specifications apply only to this filter, and not to the entire DAC. The specifications can be used to perform further analysis of the exact characteristics of the filter, for example using a digital filter design software package.)
Figure 2 shows the frequency response of the IIR low-pass filter.
Passband ripple is ±0.2 dB for the combined effects of the DAC’s digital filters (i.e., high-pass filter and IIR low pass of the interpolation filter) in the 300 Hz–3400 Hz passband.
Analog Smoothing Filter and Programmable Gain Amplifier
The programmable gain amplifier (PGA) can be used to adjust the output signal level by –15 dB to +6 dB. This gain is selected by bits 7–9 (OG0, OG1, OG2) of the AD28msp02’s control register.
The AD28msp02’s analog smoothing filter consists of a 2ndorder Sallen-Key continuous-time filter and a 3rd-order switched capacitor filter. The Sallen-Key filter has a 3 dB point at approximately 80 kHz.
Differential Output Amplifier
The AD28msp02’s analog output (VOUTP, VOUTN) is produced by a differential output amplifier. The differential amplifier can drive loads of 2 kΩ or greater and has a maximum differential output voltage swing of ±3.156 V peak-to-peak (3.17 dBm0). The output signal is dc-biased to the AD28msp02’s on-chip voltage reference (VREF) and can be ac-coupled directly to a load or dc-coupled to an external amplifier. Refer to “Analog Output” in the “Design Considerations” section of this data sheet for more information.
The VOUTP–VOUTN outputs must be used as differential outputs; do not use either as a single-ended output.
The AD28msp02 communicates with a host processor via the bidirectional synchronous serial port (SPORT). The SPORT is used to transmit and receive digital data and control information.
All serial transfers are 16 bits long, MSB first. Data bits are transferred at the serial clock rate (SCLK). SCLK equals the master clock frequency divided by 5. SCLK = 2.6 MHz for the master clock frequency MCLK = 13.0 MHz.
Host Processor Interface
The AD28msp02-to-host processor interface is shown in Figure 4.
AD28msp02 |
|
Host Processor |
SDO |
|
SERIAL DATA RECEIVE |
SDOFS |
|
RECEIVE FRAME SYNC |
SCLK |
|
SERIAL CLOCK |
DATA/CNTRL |
|
FLAG |
SDI |
|
SERIAL DATA TRANSMIT |
SDIFS |
|
TRANSMIT FRAME SYNC |
|
|
|
Figure 4. AD28msp02-to-Host Processor Interface
Table I describes the SPORT signals and how they are used to communicate with the host processor. The AD28msp02’s chip select (CS) must be held high to enable SPORT operation. CS can be used to 3-state the SPORT pins and disable communication with the host processor.
To use the ADSP-2101 or ADSP-2111 as host DSP processor for the AD28msp02, the following connections can be used (as shown in Figure 5):
AD28msp02 Pin |
|
ADSP-2101/2111 Pin |
||
|
|
|
|
|
SCLK |
– |
SCLK0 |
||
SDO |
– |
DR0 |
||
SDOFS |
– |
RFS0 |
||
SDI |
– |
DT0 |
||
SDIFS |
– |
TFS0 |
||
DATA/ |
CNTRL |
|
– |
FO (Flag Output) |
|
|
|
|
|
Table I. SPORT Signals
Signal |
|
|
Signal State When |
Signal State During |
||
Name |
Description |
Generated By |
|
RESET |
Low (CS High) |
Powerdown (CS High) |
|
|
|
|
|
|
|
SCLK |
Serial clock |
AD28msp02 |
|
Low |
Active |
|
SDO |
Serial data output |
AD28msp02 |
|
Low |
Active* |
|
SDOFS |
Serial data output frame sync |
AD28msp02 |
|
Low |
Low |
|
SDI |
Serial data input |
Host Processor |
— |
— |
||
SDIFS |
Serial data input frame sync |
Host Processor |
— |
— |
||
|
|
|
|
|
|
|
(CS must be held high to enable SPORT operation.)
*Outputs last data value that was valid prior to entering powerdown.
–4– |
REV. 0 |
AD28msp02
Note that the ADSP-2101’s SPORT0 communicates with the AD28msp02’s SPORT while the ADSP-2101’s Flag Output
(FO) is used to signal the AD28msp02’s DATA/CNTRL input. SPORT1 on the ADSP-2101 must be configured for flags and interrupts in this system.
Figure 6 shows an ADSP-2101 assembly language program that initializes the AD28msp02 and implements digital loopback through the DSP processor.
AD28msp02 SDO |
|
DR0 |
ADSP-2101 |
SDOFS |
|
RFS0 |
|
SCLK |
|
SCLK0 |
|
DATA/CNTRL |
|
FO |
|
SDI |
|
DT0 |
|
SDIFS |
|
TFS0 |
|
|
|
|
|
Figure 5. AD28msp02-to-ADSP-2101 Interface
{This ADSP-2101 program initializes the AD28msp02 }
{and executes a loopback, or talk-through, routine. }
.MODULE/ABS = 0/BOOT = 0 test1;
resetv: |
JUMP begin; |
{restart} |
|
RTI; RTI; RTI; |
|
irq2v: |
RTI; RTI; RTI; RTI; |
{IRQ2} |
st0x: |
RTI; RTI; RTI; RTI; |
{SPORT0 Tx} |
sr0x: |
ax0 = rx0; |
{SPORT0 Rx} |
|
tx0 = ax0; |
|
|
RTI; RTI; |
|
irq1v: |
RTI; RTI; RTI; RTI; |
{irq1} |
irq0v: |
RTI; RTI; RTI; RTI; |
{irq0} |
timerv: |
RTI; RTI; RTI; RTI; |
|
begin: |
RESET FLAG OUT; |
|
|
AX0 = 0x2A0F; |
{Configure ADSP-2101 SPORT0 for } |
|
DM (0x3FF6) = AX0; |
{ ext. SCLK, ext. RFS, int. TFS } |
|
AX0 = 0x101F; |
{ Enable ADSP-2101 SPORT0, } |
|
DM (0x3FFF) = AX0; |
{ configure SPORT1 for Flag Out } |
|
IMASK= 0x10; |
|
|
AX0 = 0x30; |
{ Write control word to take} |
|
TX0 = AX0; |
{ AD28msp02 out of powerdown } |
|
IDLE; |
|
|
NOP; |
|
|
IMASK= 0x08; |
|
|
SET FLAG OUT; |
|
wait: |
JUMP wait; |
{ Wait for receive interrupt } |
NOP;
.ENDMOD;
Figure 6. ADSP-2101 Digital Loopback Routine
REV. 0 |
–5– |
AD28msp02
Serial Data Output
The AD28msp02’s SPORT will begin transmitting data to the host processor at an 8 kHz rate when the PWDD and PWDA bits (Bits 4, 5) of the control register are set to 1. In the program shown in Figure 6, the instructions
AX0 = 0x30; { Write control word to take } TX0 = AX0; { AD28msp02 out of powerdown }
accomplish this by writing 0x30 to the AD28msp02’s control register. There is a short start-up time (after the end of this control register write) before the AD28msp02 raises SDOFS and begins transmitting data; see Figure 11.
At the 13 MHz MCLK frequency, data is transmitted at an 8 kHz rate with a single 16-bit word transmitted every 125 μs.
While data is being output, the AD28msp02 asserts SDOFS at an 8 kHz rate. Each 16-bit word transfer begins one serial clock cycle after SDOFS is asserted.
Serial Data Input
The host processor must initiate data transfers to the AD28msp02 by asserting the serial data input frame sync (SDIFS) high. The 16-bit word transfer begins one serial clock cycle after SDIFS is asserted. The DATA/CNTRL line must be driven high when SDIFS is driven high.
The host processor must assert SDIFS shortly after the rising edge of SCLK and must maintain SDIFS high for one cycle. Data is then driven from the host processor (to the SDI input) shortly after the rising edge of the next SCLK and is clocked into the AD28msp02 on the falling edge of SCLK in that cycle.
Each bit of a 16-bit data word is thus clocked into the AD28msp02 on the falling edge of SCLK (MSB first).
If SDIFS is asserted high again before the end of the present data word transfer, it is not recognized until the falling edge of SCLK in the last (LSB) cycle.
(Note: Exact SPORT timing requirements are defined in the “Specifications” section of this data sheet.)
The AD28msp02’s control register configures the device for various modes of operation including ADC and DAC gain settings, ADC input mux selection, filter bypass, and powerdown. The AD28msp02’s host processor can read and write to the control register through the AD28msp02’s serial port (SPORT) by driving the DATA/CNTRL pin low.
The control register is cleared (set to 0x0000) when the AD28msp02 is reset.
Control Register Writes
To write the control register, the host processor must assert DATA/CNTRL low when it asserts SDIFS. If the MSB of the bit stream is also low, the SPORT recognizes the incoming serial data as a new control word and copies it to the
AD28msp02’s control register. The format for the control word write is shown in Table II; reserved Bits 10-15 must be set to zero.
|
|
|
|
|
|
|
|
|
|
|
|
|
Table II. Control Word Write Format |
|
|
|
|
|
|||||||
15 |
|
14 |
13 |
|
12 |
11 |
|
10 |
9 |
|
8 |
|
7 |
|
6 |
5 |
4 |
3 |
2 |
1 |
0 |
||||
|
0 |
|
0 |
|
0 |
|
|
0 |
0 |
|
0 |
OG2 |
OG1 |
|
OG0 |
|
0 |
PWDD |
PWDA |
ADBY |
DABY |
IMS |
IPS |
||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
0 |
|
|
IPS |
|
Analog input preamplifier select: 1 = insert (+20 dB), 0 = bypass (0 dB) |
|
|
|
|
||||||||||||||||
1 |
|
|
IMS |
|
Analog input multiplexer select: 1 = AUX input, 0 = NORM input |
|
|
|
|
||||||||||||||||
2 |
|
|
DABY |
|
DAC high-pass filter bypass select: 0 = insert, 1 = bypass |
|
|
|
|
|
|||||||||||||||
3 |
|
|
ADBY |
|
ADC high-pass filter bypass select: 0 = insert, 1 = bypass |
|
|
|
|
|
|||||||||||||||
4 |
|
|
PWDA |
|
Powerdown analog: 0 = powerdown, 1 = operating |
|
|
|
|
|
|
||||||||||||||
5 |
|
|
PWDD |
|
Powerdown digital: 0 = powerdown, 1 = operating |
|
|
|
|
|
|
||||||||||||||
7–9 |
|
OG2-OG0 |
|
Analog output gain setting (for D/A output PGA) |
|
|
|
|
|
|
|
||||||||||||||
10–15 |
|
|
|
|
|
|
Reserved |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Gain |
|
|
|
|
|
OG2 |
|
|
OG1 |
|
|
OG0 |
|
|
|
|
|
|
|
|
|
||||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
+6 dB |
|
|
|
|
|
0 |
|
|
|
0 |
|
|
0 |
|
|
|
|
|
|
|
|
|
|
||
+3 dB |
|
|
|
|
|
0 |
|
|
|
0 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
||
0 dB |
|
|
|
|
|
0 |
|
|
|
1 |
|
|
0 |
|
|
|
|
|
|
|
|
|
|
||
–3 dB |
|
|
|
|
|
0 |
|
|
|
1 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
||
–6 dB |
|
|
|
|
|
1 |
|
|
|
0 |
|
|
0 |
|
|
|
|
|
|
|
|
|
|
||
–9 dB |
|
|
|
|
|
1 |
|
|
|
0 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
||
–12 dB |
|
|
|
|
1 |
|
|
|
1 |
|
|
0 |
|
|
|
|
|
|
|
|
|
|
|||
–15 dB |
|
|
|
|
1 |
|
|
|
1 |
|
|
1 |
|
|
|
|
|
|
|
|
|
|
|||
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Gain settings are accurate within ±0.6 dB.
(Control Register is set to 0x0000 at RESET. Reserved Bits 10–15 must be set to 0 for all Control Register writes.)
–6– |
REV. 0 |