Minimum 50 dB Missing Tone Rejection for
DMT Signals
D
Integrated TX/RX Filters
D
Integrated Digital Phase Lock Loop (DPLL)
and VCXO DAC
D
Integrated Equalizer for Receive Channel
D
Integrated PGA in Receive, and PAA in
Transmit Channels
description
The TLFD500PN is a high-speed analog front end for a remote terminal-side ADSL G.Lite modem. The device
is designed to perform transmit encoding (D/A conversion), receive decoding (A/D conversion), transmit and
receive filtering functions, and receive equalizer functions for a frequency division multiplex (FDM) G.Lite
application. The receive channel has an update rate of 1.104 Msps, while the transmit channel has an update
rate of 276 ksps. Both channels use 2s complement data format.
D
Direct Single Serial Interface to TI’s C54x or
C6x DSP (Data and Control)
D
Eight General-Purpose I/O Pins
D
Software and Hardware Power-Down
Modes
D
Industrial Temperature Range
(–40°C to 85°C)
D
Integrated Auxiliary Amplifiers for System
Flexibility
D
Single 3.3 V Supply
D
80-Pin LQFP (PN) Package
D
2s Complement Data Format
When used in a G.Lite system, the TLFD500PN requires a minimum number of external components. The
device incorporates integrated filtering, DPLL, VCXO DAC (uses 2s complement data format), and 8
general-purpose I/O ports. The general-purpose I/O ports provide a means of reading or writing status bits in
the system. Four auxiliary amplifiers on the chip can be configured (external components may be required) to
provide additional onboard filtering and amplification.
A simple serial interface for data transfer on the digital side reduces system component count. The interface
can be connected directly to the TI C6x and C54x families of DSPs.
The TLFD500PN device is available in an 80-pin PN LQFP package.
Please be aware that an important notice concerning availability, standard warranty, and use in critical applications of
Texas Instruments semiconductor products and disclaimers thereto appears at the end of this data sheet.
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of Texas Instruments
standard warranty. Production processing does not necessarily include
testing of all parameters.
AMP1INP– AMP4INP11,2,66,59IAuxiliary amplifier 1–4 positive input
AMP1INM– AMP4INM10,3,65,60IAuxiliary amplifier 1–4 negative input
AMP1OUTP– AMP2OUTP9,4OAuxiliary amplifier 1–2 positive output. Outputs are self-biased to AVDD_TX/2.
AMP3OUTP– AMP4OUTP64, 61OAuxiliary amplifier 3–4 positive output. Outputs are self-biased to AVDD_RX/2.
AMP1OUTM– AMP2OUTM12,1OAuxiliary amplifier 1–2 negative output. Outputs are self-biased to AVDD_TX/2.
AMP3OUTM– AMP4OUTM67,62OAuxiliary amplifier 3–4 negative output. Outputs are self-biased to AVDD_RX/2.
AVDD_REF47IAnalog supply for reference circuit
AVDD_RX48,68IRX channel analog supply
AVDD_TX7ITX channel analog supply
AVSS_REF44IAnalog supply return for reference(analog ground)
AVSS_RX49,69IRX channel analog supply return (analog ground)
AVSS_TX8ITX channel analog supply return (analog ground)
COMPDAC116ITX channel decoupling cap input A. Add 1 µF capacitor to AVDD_TX
COMPDAC215ITX channel decoupling cap input B. Add 1 µF capacitor to AVDD_TX
DGPO35ODirect general-purpose output. This pin reflects the last value written to the DGPO bit
56,57
FSX22OSerial port frame sync transmit signal
FSR21OSerial port frame sync receive signal
GPIO0–GPIO723–30I/OGeneral-purpose I/O
HPF1INP78IRX channel stage 1 amplifier positive input. Input signal needs to have AVDD_RX/2
HPF1INM77IRX channel stage 1 amplifier negative input. Input signal needs to have AVDD_RX/2
HPF2INP74IRX channel stage 2 positive input. Input signal need to have AVDD_RX/2 common mode
HPF2INM73IRX channel stage 2 negative input. Input signal need to have AVDD_RX/2 common mode
HPF1OUTP76ORX channel stage 1 amplifier positive output. Used to connect external components to obtain
HPF1OUTM79ORX channel stage 1 amplifier negative output. Used to connect external components to
HPF2OUTP72ORX channel stage 2 positive output. Output signal has AVDD_RX/2 common mode voltage.
HPF2OUTM75ORX channel stage 2 negative output. Output signal has AVDD_RX/2 common mode voltage.
MCLKIN/PLLCLKIN37IMultiplexed pin based on value of PLLSEL. Selects master clock input, or clock input for PLL
location in the SDR data stream. It is a general-purpose output that does not require a
secondary transfer to control.
IRX channel digital supply return (digital ground)
common mode voltage.
common mode voltage.
voltage.
voltage.
stage 1 HPF.
obtain stage 1 HPF.
mode.
4
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
I/O
DESCRIPTION
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
Terminal Functions (Continued)
TERMINAL
NAMENO.
NC36,53, 58,
63
PLLSEL42ISelects between VCXO mode and DPLL mode. If the pin is tied high PLL mode is selected.
PWRDN17IPower-down pin. When PWRDN is pulled low the device goes into power-down mode. The
REFM45ONegative reference filter node. This terminal is provided for low-pass filtering of the internal
REFP46OPositive reference filter node. This terminal is provided for low-pass filtering of the internal
RESET38IDevice reset input pin. Initializes all the device’s internal registers to their default values.
RXBANDGAP43ORX channel band-gap filter node. This terminal is provided for decoupling of the 1.5-V
RXINP70IRX channel stage 3 positive input. The input is self-biased at AVDD_RX/2.
RXINM71IRX channel stage 3 negative input. The input is self-biased at AVDD_RX/2.
SCLK19OSerial port shift clock (transmit and receive)
SDR20ISerial data receive from DSP
SDX18OSerial data transmit to DSP
TXBANDGAP14OTX channel band-gap filter node. This terminal is provided for decoupling of the 1.5-V
TXOUTP5OTX channel positive output
TXOUTM6OTX channel negative output
VCXOCNTL13ODAC output to control onboard VCXO
VMID_RX50I/ODecoupling Vmid for ADC. Add 10 µF (tantalum) and 0.1 µF (ceramic) capacitors to analog
VSS80ISubstrate. Connect to analog ground.
No connection. Keep floating.
Pin should be tied low for VCXO mode. Cannot be left floating.
default state of this pin is low.
band-gap reference. The optimal ceramic capacitor value is 10 µF (tantalum) and 0.1 µF
(ceramic), connected to analog ground. The nominal dc voltage at this terminal is 0.5 V.
band-gap reference. The optimal ceramic capacitor value is 10 µF (tantalum) 0.1 µF
(ceramic), connected to analog ground. The nominal dc voltage at this terminal is 2.5 V.
The default state of this pin is low.
band-gap reference. The optimal capacitor value is 10 µF (tantalum) and 0.1 µF (ceramic).
This node should not be used as a voltage source.
band-gap reference. The optimal capacitor value is 10 µF (tantalum) and 0.1 µF (ceramic).
This node should not be used as a voltage source.
ground.
detailed description
transmit
The transmit channel is powered by a high performance DAC. The transmit channel update rate is 276 kHz.
The DAC is a 14-bit DAC at 4.416-MHz. This provides 16X oversampling. A band-pass filter limits the output
of the transmitter to a frequency range of 30 kHz to 138 kHz. A differential amplifier drives the output into the
external line driver. The dif ferential amplifier has programmable attenuation for added flexibility. The transmitter
high-pass filter can be bypassed by writing the appropriate bit to the filter bypass control register (BCR).
The output spectrum of the DAC complies with the nonoverlapped power spectrum density (PSD) mask
specified in the ITU draft recommendation G.992.2 for G.Lite.
The TXP AA is a programmable-attenuation amplifier. It provides 0 dB to 24 dB of attenuation in1-dB steps. The
TXP AA is controlled via the P AA control register (PCR). For details about register programming see the register
programming section.
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
5
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
detailed description (continued)
receive
The receive channel consists of a high-pass filter, a programmable gain amplifier, an ADC, and filters. In
addition, it has an equalizer to attain maximum system performance. The input of the receiver is fully differential.
The ADC in the receive channel is a 14-bit converter which samples at 4.416 Msps for 4X oversampling. An
on-chip decimator reduces the sampling frequency to 1.104 MHz. The low pass filtering of the receive channel
limits the converted data to frequencies below 552 kHz.
The high-pass analog filter is used to reject the near-end echo to maximize the dynamic range of the ADC. The
high-pass filter consists of two stages: (1) a second order high-pass filter (HPF1) and, (2) a third order elliptic
high-pass filter (HPF2). Both stages have a cutoff at 180 kHz. The filter is divided into two stages to minimize
the noise from a single stage being amplified throughout. T ogether , the two high-pass filters typically attenuate
the echo power by 30 dB. There is a programmable gain amplifier (PGA) between the two filters for coarse gain
adjustments of 0-dB –12-dB in 3-dB steps. After the high-pass filter stage, the receiver channel has a
0-dB –18-dB PGA that can be adjusted in 6-dB steps. HPF2 and PGAs are integrated in one block. Figure 1(a),
1(b), and 1(c) show the frequency response of HPF1 and HPF2 (with PGAs).
The PGA is followed by a 552-kHz low-pass filter with a programmable 25-dB/MHz slope (5-dB/MHz step)
equalizer incorporated. After the equalizer, there is a fine-gain adjustment PGA of 0-dB to 9-dB in 0.25-dB steps.
All the RX PGAs are controlled via the PGA control registers (PCR–RX1 and PCR–RX2). See the register
programming section for details about register programming.
6
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
detailed description (continued)
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
50
0
–50
–100
Gain – dB
–150
–200
–250
00.51
f – Frequency – Hz
(a) RX-Stage HPF1 Frequency Response (0 to 200 kHz)
0
–
3
–
6
–
9
–12
dB
–15
–18
–21
–24
–27
1.52
× 10
2
0
–2
–4
Gain – dB
–6
–8
–10
5
11.21.4
f – Frequency – Hz
(b) RX-Stage HPF1 Frequency Response (100 kHz to 200 kHz)
Figure 1. RX Stage HPF1 and HPF2 Frequency Response
clock control – VCXO mode
The VCXODAC uses a 12-bit, 2s complement number to control a 0-V to 3-V analog output. The two 8-bit
registers, VCR-M and VCR-L, are used to generate the 12-bit control code (2s complement). This implies the
use of 16 bits to obtain a 12-bit number.
VCR-M register occupy the most significant 8 bits in the 12-bit number and the lower 4 bits of the VCR-L register
(VCR-L[3:0] ) are used for the low 4 bits of the 12-bit number. The 12-bit code is updated every time either
register is updated. VCR-L[7:4] must always be zero.
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
7
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
clock control – DPLL mode
As an alternative to the VCXODAC and VCXO, an off-chip crystal oscillator (XO) followed by an on-chip digital
PLL are also implemented. Refer to Figure 7 for an internal function block diagram. The input clock (35.328
MHz) goes to a programmable frequency divider to generate sampling clock for the ADC and DAC converters.
By changing the divide ratio, the phase of the sampling clock can be adjusted. Setting PLLSEL (pin 42) high
will enable the DPLL mode. Refer to DPLL section for detail.
clock generation
The clock generation block creates the necessary internal and external clocks needed by the device. All the
clocks generated are produced from the CLKIN signal.
The following are recommended operational parameters for the external VCXO:
3.3-V supply, 35.328 MHz ±50 PPM center frequency , and input control voltage range of 0 V–3 V.
The recommended duty cycle is 50/50.
clock generation – SCLK
SCLK is an output and is used for serial data transfer. It runs at 35.328 MHz. Although SCLK and MCLK run
at the same speed, there is no fixed phase relationship between them.
serial interface
The serial interface on the TLFD500PN connects directly to TI’s C54x or C6x families of DSPs. The interface
operates at 35.328 MHz. The serial port consists of five signals: SCLK, FSX, FSR, SDX, and SDR. A typical
connection diagram is shown in Figure 2.
DSP
CLKR
CLKX
FSX
FSR
DX
DR
TLFD500PN
SCLK
FSR
FSX
SDR
SDX
Figure 2. Typical Serial Port Connection
The serial port utilizes a primary/secondary scheme to transfer conversion data and control register data. A
primary transfer scheme, used to transfer conversion data, occurs every conversion period. A secondary
transfer scheme, used to transfer control data, happens only when requested by the host processor. The host
processor requests a secondary transfer by using the LSB of the SDR data of the primary scheme. A value of
1 indicates a secondary transfer request. Once the secondary request is made and the primary transfer has
been completed, secondary frame sync pulse (FSX/FSR) are transmitted to the host processor to indicate the
beginning of the secondary transfer. The secondary FSX signal arrives 16 SCLKs after the primary FSX, and
thus 48 SCLKs after the host processor request. This is because the span between FSX pulses for primary
transfers is always 32 SCLKs. Each bit is read/written at the rising edge of the SCLK clock. Data bit mappings
and example data transfers are shown in Table 1.
8
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
detailed description (continued)
Table 1. SDR LSB Control Function
CONTROL BIT D0CONTROL BIT FUNCTION
0No secondary transfer requested
1Secondary transfer requested
primary transfer data mapping
The data bit mapping of a primary transfer is shown in Figure 3. Bits D2–D15 of the SDR data stream are DAC
data. D1 is the control bit for the DGPO pin. The value written to this bit is reflected on the DGPO pin. See the
timing diagram in Figures 5 and 6 for detailed timing information. D0 is the secondary transfer request bit. When
a 1 is written to this bit, the host is requesting a secondary data transfer.
In the SDX data stream, D2–D15 contain the ADC conversion data. D0 and D1 can be set to reflect the values
of GPIO1 and GPIO2. To set D0 and D1 to reflect the GPIO values, the proper bit in the MCR register needs
to be set.
Secondary
Transfer
Data to CODEC
DGPO Bit
Request
SDR
SDX
GPIOx Status if Configured as
Input. Zero if GPIOx Configured
Data from CODEC
D15 – D2
D1D0D15 – D2
as output or if Masked Off
GPIO1GPIO0
Figure 3. Primary Transfer Data Bit Mapping
secondary transfer data mapping
Secondary serial communication is used to configure the device. The data bit mapping for a secondary transfer
is shown in Figure 4. Bits D10–D14 of the SDR data from the host contain the address of the control register
involved in the transfer. D15 is a R/W bit. To read out the control register by the host processor, bit R/W must
be set to 1. T o write to the control register by the host processor, bit R/W must be set to 0. During a read operation,
bits D0–D7 are don’t care. For a write operation, bits D0–D7 contain the data for the register addressed by
D10–D14. The eight bits of SDX always reflect the status of GPI00–7.
If the secondary transfer is a read operation, the contents of the control register addressed by D10–D14 of the
SDR data are reflected in bits D0–D7 of the SDX data stream. If the secondary transfer is a write operation, bits
D0–D7 on SDX will be all zeroes.
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
9
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
secondary transfer data mapping (continued)
D9D0
A0
SDR (Read)
SDX (Read)
D15
A4A3A2A1
1
Register AddressDon’t Care
D15D0D8D7
GPIO0 – 7 Status
Read Cycle (Codec Register Data Read by DSP)
D8D7
Don’t Care
Register Data
SDR (Write)
SDX (Write)
D9D8 D7
0
A4A3A2
Register Address
D15D0D8D7
GPIO0 – 7 Status
A1A0
Don’t Care
Write Cycle (DSP Data Write to Codec Register)
Data to the Register
All 0
D0
Figure 4. Secondary Transfer Data Bit Mapping
example data transfers
Figures 5(a) and 5(b) show the timing relationship for SCLK, FSX, SDX, FSR, and SDR in a primary
communication. The timing sequence for this operation is as follows:
1.FS is set high and remains high during one SCLK period, then returns to low.
2.A 16-bit word is transmitted from the ADC (SDX), and a 16-bit word is received for DAC conversion (SDR).
Figure 6(a) and 6(b) shows the timing relationship with secondary request.
10
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
detailed description (continued)
SCLK
(Output)
FSX
(output)
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
SDX
(Output)
SCLK
(Output)
FSR
(output)
SDR
(Input)
D15D14D13 D12D11 D10D9D8D7D6D5D4D3D2D1D0
t1 t2 t3
t1 t2t3
t1: DSP detects FSX
t2: TLFD500PN sends data
t3: DSP latches data
D15
D14 D13D12D11D10D9D8D7D6D5D4D3D2D1D0
t1: DSP detects FSR
t2: DSP sends data
t3: TLFD500PN latches data
(a) TLFD500PN to DSP
(b) DSP to TLFD500PN
NOTE: TI DSP requires 10 ns after the positive edge of the SCLK to give the SDR data. This plus the board delay, output buffer (for SCLK) and
input buffer delay (for SDR) to around 17 ns. As a consequence the SDR data can not be latched at the negative edge of SCLK.
Figure 5. Data Transfers
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
11
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
detailed description (continued)
128 SCLKs
FSR
FSX
16 SCLKs
SDR
SDX
FSR
P
48 SCLKs
DataCommand
DataDataDataDataData
P
Don’t CareDon’t Care
Zeroes
S
Status
(a) With Secondary Request
128 SCLKs
ZeroesZeroes
P
PPPPPS
Data
P
FSX
16 SCLKs
SDR
SDX
32 SCLKs
PPPPP
Data
DataDataDataDataData
Zeroes
Zeroes
(b) Without Secondary Request
Don’t Care
ZeroesZeroes
Data
Figure 6. Data Transfers
12
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
detailed description (continued)
general purpose I/O port (GPIO)
The general-purpose I/O port provides eight input/output pins and one output-only pin for control of external
circuitry , or for reading the status of external devices. The eight input/output pins are labeled GPIO0 –GPIO7.
The output-only pin is labeled DGPO (direct general-purpose output). This pin is labeled as direct because a
secondary transfer is not required to write to this pin.
The GPIO pins are controlled and read in the GPR-D register. The GPR-C register is used to configure the GPIO
pins as input or output pins. The default reset condition is 1 1111111b, indicating that all are configured as inputs.
For further details on register programming see the register programming section. The DGPO pin does not need
configuring and is controlled by the D1 bit in the SDR data stream (that is, from the DSP to the TLFD500PN)
during primary data transfers. In addition, a secondary transfer is not required to read GPIO0 and GPIO1 when
they are configured as inputs. Their values can be mapped into the lower two bits of the SDX data stream (that
is, from TLFD500PN to DSP) during primary data transfers. To map the values of GPIO0 and GPIO1 into the
lower two bits of the SDX ADC data stream, set the appropriate bit in the MCR register.
For more flexibility, the values of GPIO0 – GPIO7 are mapped into the upper eight data bits of the SDX data
stream on secondary data transfers. This allows the host processor to read the values of the GPIO pins and
the contents of another control register during the same secondary data transfer. When a GPIO pin is being
configured as an output, its corresponding status bit in the SDX data stream will be the last value written to the
output pin.
Each output is capable of driving 2 mA.
reference system
The integrated reference provides voltage and current to the internal analog blocks. It is also brought out to
external pins for noise decoupling. They should not be used as dc voltage source.
When the internal reference is being used by the device, the device may be powered down by writing the
appropriate reference control bit in the main control register (MCR) to achieve power savings during periods
of device inactivity.
auxiliary amplifiers
Four auxiliary high-performance operational amplifiers on the chip allow for additional onboard filtering and
amplification with minimal component count. Each op-amp has differential inputs and outputs, with 2 input pins
and 2 output pins. Each op-amp can be enabled by register programming.
The typical specifications for the operational amplifiers are as follows:
DC Gain:126 dB
Bandwidth:116 MHz
PSRR:100 dB at dc, 70 dB at 1 MHz, and 40 dB at 4 MHz
Output common-mode: AVDD_RX/2 (auxiliary amplifier 3,4) or AVDD_TX/2 (auxiliary amplifier 1,2)
Input interface:AC coupled
device power-up sequence
All digital and analog supplies must be properly biased. All supply pins are mandatory . The power supply can
not be switched, even when the codec has been powered down or parts of the codec are in power-down mode.
Reset must be held at least 20 µs after power up. To reset the reference circuit and registers requires 100 ms.
When the chip is woken up from hardware power-down mode, it takes100 ms to reset the reference circuit
before the chip works in normal mode. When the chip is woken up from software power-down mode, only 20
µs is needed before valid data comes out (reference must be kept on). Register values will not change in either
wake-up operation.
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
13
TLFD500PN
NAME
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
detailed description (continued)
register programming
The codec registers are listed in Table 2, with each bit of each register defined. All registers are 8-bit wide.
NOTE:
Bits not defined in the table are reserved for future use. During a read, the reserved bit read value
is not guaranteed. During a write, only zeroes can be written to reserved bits.
Table 2. Codec Registers
REGISTER
ADDRESS
A4 A3 A2 A1 A0
BCR00001R/WD0: Power-down RX HP filter 1
PCR-RX100010R/WD[5:0] = RXPGA3[5:0]; Fine gain, 0 to 9 dB, 0.25-dB steps
PCR-RX200011R/WD[2:0] = RXPGA1[2:0]; 0 to 12dB, 3-dB steps
PCR-TX00100R/WD[4:0] = TX PAA[4:0]; 0 to –24dB, –1-dB steps
EQR00101R/WD[2:0] = EQ[2:0] 0 to 25 dB, 5 dB/MHz steps; D[6:4] = EQ_PGA[2:0] 0 to 6 dB, 1 dB
VCR-M00110R/WD[7:0] = VCXO DAC control Bit[11:4].
VCR-L00111R/WD[3:0] = VCXO DAC control Bit[3:0]. D[7:4] must always be zero.
GPR-C01000R/WD[7:0] = GPIO1 I/O control (0 = output, 1 = input)
GPR-D01001R/WD[7:0] = GPIO data register
Reserved01010R/WFor future use. Read or write of register not allowed.
AUXR0101 1R/WD0: Enable auxiliary amplifier 2
NCO_DEF01100R/WD[7:0] = Default NCO divide number
NCO_DIV_DELAY01 101R/WD[7:0] = Number of samples, from current secondary transfer, after which effect of delta
NCO_DELTA01110R/WD[7:4] = Delta from default for first sample of data frame (–8 through 7)
MCR01111R/WD0: S/W Power-down main reference
MODEFUNCTION
D1: Power-down RX HP filter 2
D2: Bypass TX digital HP filter
D3: Echo mode: Echo SDR data back to SDX
D4: Reserved
D5: Reserved
D6: Reserved
D[3:0] = Number of times NCO divider remains changed from default before being set
back to default (0 through 15)
D1: S/W Power-down TX channel with reference still on
D2: S/W Power-down RX channel with reference still on
D3: S/W Power-down VCXO with reference still on
D4: S/W Reset
D5: Analog loop back (refer to block diagram)
D6: Digital loop back (refer to block diagram)
D7: Enable GPIO 1 and 2 to show in SDX primary data
R––––––––ReservedBit reserved for future use
–R–––––––ReservedBit reserved for future use
––R––––––ReservedBit reserved for future use
–––R–––––ReservedBit reserved for future use
––––0––––
––––1–––0x08
–––––0–––
–––––1––0x04
––––––0––
––––––1–0x02
–––––––0–
–––––––10x01
NOTE 1: ECHO mode allows for a quick verification of the serial interface operation. It sends back the data from input data buffer to the output
data buffer and does not exercise the RX or TX channel.
D6D5D4D3D2D1D0
Table 3. EQR Bit Definition
Do not echo SDR data on SDX
Echo SDR data on SDX (see Note 1)
Enable TX HP Filter
Bypass TX HP Filter
Power up RX HP Filter 2
Power down RX HP Filter 2
Power up RX HP Filter 1
Power down RX HP Filter 1
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
15
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
PCR-RX1 – programmable gain control register 1 for RX channel PGA3
NOTE 2: The formula to convert bit value to RXPGA3 gain in dB is
RXPGA3 gain (in dB) = RXPGA3[5:0] (in decimal) x 0.25dB
Similarly one can compute the RXPGA3 [5:0] bit combination needed, given the gain in dB.
CAUTION:
Performance of the codec for invalid combination of bits is not guaranteed and such
combinations should not be used. The user should make no assumption that the code bits
will saturate to a maximum or minimum value or wrap around to a valid combination.
TLFD500PN
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
17
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
PCR-RX2 – programmable gain control register 2 for RX channel PGA1 and PGA2
–R–––––––ReservedBit reserved for future use
––R––––––ReservedBit reserved for future use
–––00–––0x00RXPGA2[1:0]0dB
–––01–––0x086dB
–––10–––0x1012dB
–––11–––0x1818dB
–––––0000x00RXPGA1[2:0]0dB
–––––0010x013dB
–––––0100x026dB
–––––0110x039dB
–––––1000x0412dB
–––––1010x05INVALID
–––––1100x06INVALID
–––––1110x07INVALID
NOTES: 3. The formula to convert bit value to RXPGA2 gain in dB is
RXPGA2 gain (in dB) = RXPGA2[1:0] (in decimal) x 6dB
Similarly the needed RXPGA2[1:0] bit combination can be computed, given the gain in dB.
4. The formula to convert bit value to RXPGA1 gain in dB is
RXPGA1 gain (in dB) = RXPGA1[2:0] (in decimal) x 3dB
Similarly the needed RXPGA1[2:0] bit combination can be computed, given the gain in dB.
Performance of the codec for invalid combination of bits is not guaranteed and such
combinations should not be used. The user should make no assumption that the code bits
will saturate to a maximum or minimum value or wrap around to a valid combination.
D6D5D4D3D2D1D0
T able 5. PCR-RX2 Gain
CAUTION:
PCR-TX – programmable attenuation control register for TX channel
PCR-TX – programmable attenuation control register for TX channel (continued)
Table 6. PCR-TX Attenuation
D7D6D5D4D3D2D1D0HEX VALUEBIT NAMEDESCRIPTION
R––––––––ReservedBit reserved for future use
–R–––––––ReservedBit reserved for future use
––R––––––ReservedBit reserved for future use
–––000000x00TXPAA[4:0]–0 dB
–––000010x01–1 dB
–––000100x02–2 dB
–––000110x03–3 dB
–––001000x04–4 dB
–––001010x05–5 dB
–––001100x06–6 dB
–––001110x07–7 dB
–––010000x08–8 dB
–––010010x09–9 dB
–––010100x0A–10 dB
–––010110x0B–11 dB
–––011000x0C–12 dB
–––011010x0D–13 dB
–––011100x0E–14 dB
–––011110x0F–15 dB
–––100000x10–16 dB
–––100010x11–17 dB
–––100100x12–18 dB
–––100110x13–19 dB
–––101000x14–20 dB
–––101010x15–21 dB
–––101100x16–22 dB
–––101110x17–23 dB
–––110000x18–24 dB
–––110010x19INVALID
–––110100x1AINVALID
–––110110x1BINVALID
–––111000x1CINVALID
–––111010x1DINVALID
–––111100x1EINVALID
–––111110x1FINVALID
NOTE 5: The formula to convert bit value to TXP AA attenuation in dB is
TXPAA attenuation (in dB) = TXPAA[4:0] (in decimal) x (–1)dB
Similarly one can compute the TXPAA[4:0] bit combination needed, given the attenuation in dB.
CAUTION:
Performance of the codec for invalid combination of bits is not guaranteed and such
combinations should not be used. The user should make no assumption that the code bits
will saturate to a maximum or minimum value or wrap around to a valid combination.
NOTES: 6. The formula to convert bit value to EQPGA gain in dB is
EQPGA gain (in dB) = EQPGA[2:0] (in decimal) x 1 dB
Similarly one can compute the EQPGA[2:0] bit combination needed, given the gain in dB.
7. The formula to convert bit value to EQ slope in dB is
EQ slope (in dB/MHz) =EQ[2:0] (in decimal) x 5 dB/MHz
Similarly one can compute the EQ[2:0] bit combination needed, given the slope in dB/MHz.
Performance of the codec for invalid combination of bits is not guaranteed and such
combinations should not be used. The user should make no assumption that the code bits
will saturate to a maximum or minimum value or wrap around to a valid combination.
–R–––––––ReservedBit reserved for future use
––R––––––ReservedBit reserved for future use
–––R–––––ReservedBit reserved for future use
––––0––––
––––1–––0x08
–––––0–––
–––––1––0x04
––––––0––
––––––1–0x02
–––––––0–
–––––––10x01
CAUTION:
Performance of the codec for invalid combination of bits is not guaranteed and such
combinations should not be used. The default condition is with the amplifiers switched off.
NCO_DEF – numerically controlled oscillator default value register
Address: 01100bContents at reset: 01000000b (64 decimal)
111111100×FEJitter after 254 sample clocks
111111110×FFJitter after 255 sample clocks
NOTES: 9. The formula to convert NCDLY[7:0] to delay is straightforward.
Delay (number of ADCLK periods) = NDCLK[7:0] (except for 0 and 1).
10. ADCLK–A/D converter sampling clock
This register is also the only means of communicating to the codec that the ADCLK must
be jittered. Thus not writing a value implies that jitter will not take place even if other
registers have non-default values. As a side consequence, this register does not remember
its value. All the others store them unless RESET.
D6D5D4D3D2D1D0
Table 12. NCO Default Value
ADCLK jittered 2 sample clocks (of
ADCLK) after write into the
NCDLY[7:0]
CAUTION:
NCO_DIV_DELAY register (see Note
10)
Jitter after 3 to 253 sample clocks (All
individual values are valid)
Writing 0 or 1 is not recommended
examples:
1.NCDEF[7:0] = 64 (dec.), NCDEL[4:0] = 1, NCRPT[2:0] = 2, NCDLY[7:0] = 5. This shows a default division
value of 64, giving a normal ADCLK of 2.208 MHz (assuming 35.328 MHz input); the division ratio will be
64 + 1 = 65 to effect the jitter , that is, pulling in the clock phase. The jitter will be repeated for 2 consecutive
samples. The jitter will take effect 5 ADCLK sample periods after writing to NCDLY .
2.NCDEF[7:0] = 63 (decimal), NCDEL[4:0] = –1, NCRPT[2:0] = 2, NCDLY[7:0] = 5. Similar to 1. The default
frequency is slightly less than 2.208 MHz. Since the division ratio is 63 – 1 = 62, the clock phase is pushed
out.
3.NCDEF[7:0] = 64 (decimal), NCDEL[4:0] = 0, NCRPT[2:0] = 2, NCDLY[7:0] = 5. Here the jitter will not be
observed, since the delta register is zero.
4.NCDEF[7:0] = 64 (decimal), NCDEL[4:0] = 1, NCRPT[2:0] = 0, NCDLY[7:0] = 5. Here the jitter will not be
observed, since the repeat register is zero.
5.NCDEF[7:0] = 64 (decimal), NCDEL[4:0] = 1, NCRPT[2:0] = 2, NCDLY[7:0] = 0. This is invalid and not
recommended. NCDLY[7:0] can not be 0 or 1.
6.NCDEF[7:0] = 64 (decimal), NCDEL[4:0] = 1, NCRPT[2:0] = 2. Here the jitter will not occur since there was
not writing to NCDLY. The other registers will retain their values as in all other cases.
24
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
NCDEL[3:0]
NCRPT[3:0]
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
NCO_DELTA – numerically controlled oscillator delta value register
NOTES: 11. The SWRST and SWREFPD refer to the word software, since the reset is done by register programming as opposed to hard resets
done by forcing pin logic levels.
12. Analog loop-back means looping back of the analog TX output to the RX input. This way the codec can be tested without need of
external analog sources.
13. Digital loop-back means looping back the digital RX output to the TX input. Here we can test the code without the need for a DSP
and serial data transfer.
All power downs of VCXODAC, RX, and TX channels occur with the reference still on.
D6D5D4D3D2D1D0
Table 14. MCR Control
No effect on SDX
Show GPIO 1 and 2 in SDX primary.
No effect on digital loop back
Enable digital loop back
No effect on analog loop back
Enable analog loop back
No effect on reset
Perform soft reset
Power up VCXODAC
Power down VCXODAC
Power up RX channel
Power down RX channel
Power up TX channel
Power down TX channel
Power up (soft) main reference
Power down (soft) main reference
CAUTION:
DPLL detailed description
26
The default value of register NCO_DEF is 64. With the 35.328 MHz input clock, the output frequency of the PLL
is 4 × 35.328 = 141.312 MHz. To obtain an ADC clock of 2.208 MHz the divide ratio (controlled by register
NCO_DEF) needs to be 64. Increasing or decreasing this ratio (for example, 65 or 63) can effect a temporary
phase shift. The ratio is controlled by the DSP through register programming.
In DPLL mode, the ADC clock (ADCLK) will work at 2.208 MHz instead of the 4.416 MHz used in the VCXO
mode. The DAC clock (DACLK) will continue to work at 4.416 MHz. When the ADCLK is jittered, the DACLK
is also jittered.
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
DPLL detailed description (continued)
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
MCLKIN/
PLLCLKIN
PLL (X4)
NCO_DELTA
[3:0]
NCO_DEFNCO_DIV_DELAY
+
NCO_DELTA
[7:4]
CLOCK TO
CONVERTER
Figure 7. DPLL Internal Function Block Diagram
Example: Assume MCLKIN/PLLCLKIN=35.328 MHz. When NCO_DEF is programmed as 64, a 2.208 MHz
clock is provided to the ADC converter according to the following formula:
35.328 × 4/64 = 2.208
If NCO_DELTA [7:4] is set to –1, NCO_DELTA [3:0] is set to 3, and NCO_DIV_DELAY is set to 2
(NCO_DIV_DELAY should be the last register to be programmed), register NCO_DEF will change to
63,63,63,64 at the beginning of the third sampling period. Each number (63 or 64) only last one clock
(2.208 MHz) cycle. And the combination 63,63,63,64 occurs only once. Reprogramming of register
NCO_DIV_DELAY is needed if further adjustment is required.
Figure 7 shows the timing of SCLK with the following setting:
Also note that in DPLL mode, the ADC clock will work at 2.208 MHz, 2 times oversampled, (instead of 4.416 MHz
used in the VCXO mode) and the DAC clock will continue to work at 4.416 MHz.
T o prevent variation in the serial clock width (SCLK) due to jitter , the serial clock is modified to have a fixed high
level of 14 ns and a low level of 7 ns, thus resulting in a period of 21 ns. The average clock frequency is still
35.328 MHz. After the 16 SCLKs are complete, the clock goes quiet (no toggle zone) until the rising edge of
the next ADCLK. The length of the
14 ns
SCLK
35.328 MHz
7 ns
ADCLK
2.208 MHz
DACLK
4.416 MHz
no-toggle zone
varies with the ADCLK. Figure 9 illustrates an example.
1/35.328 MHz
Figure 9. Relation of SCLK With ADCLK/DACLK in DPLL Mode
absolute maximum ratings over operating free-air temperature (unless otherwise noted)
†
Supply voltage, AVDD to AGND, DVDD to DGND –0.3 V to 4.5 V. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Analog input voltage range to AGND –0.3 V to AVDD+0.3 V. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Digital input voltage range –0.3 V to DVDD+0.3 V. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Operating
Operatingfree-air temperature range, T
Storage temperature range, T
Lead temperature 1,6 mm (1/16 inch) from case for 10 seconds 250°C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
†
Stresses beyond those listed under “absolute maximum ratings” may cause permanent damage to the device. These are stress ratings only, and
functional operation of the device at these or any other conditions beyond those indicated under “recommended operating conditions” is not
implied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.
recommended operating conditions
power supply
MINNOMMAXUNIT
pp
y v
digital inputs
High-level input voltage, V
Low-level input voltage, V
NOTES: 14. The input signal is the digital equivalent of a sine wave (digital full scale = 0 dB). The normal differential output with this input condition
Gain error–1.51.5dB
PAA step gain error±0.25dB
DC offset50100mV
Cross-talkRX to TX channel–70dB
Idle channel noise65µVrms
Group delay30µs
Power supply rejection ratio (PSRR)200 mVp-p at 75 kHz70dB
Analog output voltageLoad = 2000 Ω3Vp-p
Filter gain relative to gain at 77.625 kHz
is 3 Vpp.
15. 27 tones, 25.875 to 138 kHz, 4.3125 kHz/step, 0 dB
DC offset50100mV
Cross-talkTX to RX channel–55dB
Group delay25µs
Idle-channel noise100µVrms
Common-mode rejection ratio (CMRR)70dB
Power supply rejection ratio (PSRR)200 mVp-p at 75 kHz70dB
Analog input self-bias dc voltage1.5V
timing requirements over recommended ranges of supply voltage and operating free-air
temperature (unless otherwise noted)
serial port (see Figures 7 and 8) and DGPO (see Figure 9)
PARAMETERMINTYPMAXUNIT
t
Period, SCLK28.3ns
c1
t
Delay time, FSR high before SCLK↓7ns
d1
t
Delay time, FSR high after SCLK↓7ns
d2
t
Delay time, FSX high before SCLK↓7ns
d3
t
Delay time, FSX high after SCLK↓7ns
d4
t
Delay time, SDX data valid after SCLK↑7ns
d5
t
Delay time, GPIO becomes valid after data is sent7ns
d6
t
Falling time, SCLK change from high to low4.4ns
f
t
Hold time, SDR keep valid after SCLK↑2ns
h1
t
Rising time, SCLK change from low to high4.6ns
r
t
Setup time, SDR valid before SCLK↑6ns
su1
SCLK
(Output)
FSR
(Output)
SDR
(Input)
SCLK
(Output)
FSX
(Output)
SDX
(Output)
t
c1
t
d1
t
su1
t
d2
t
h1
D15D14D13D12D11
t
r
t
f
Figure 10. Data Transfers From DSP to TLFD500PN
t
d3
t
d4
t
d5
32
Figure 11. Data Transfers From TLFD500PN to DSP
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
td6 = 7 ns
SCLK
50
0
–50
Gain – dB
–100
–150
GPIO
Valid Data
Figure 12. GPIO Bit-to-Pin Update Timing
TRANSMIT CHANNEL RESPONSE
–200
00.20.40.6
f – Frequency – MHz
Figure 13. Transfer Characteristic of the Transmit Filters
(Complies with ITU G.992.2 PSD requirement)
0.81
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
33
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
TRANSMIT CHANNEL RESPONSE WITH HP FILTER BYPASSED
50
0
–50
Gain – dB
–100
–150
–200
00.20.40.6
f – Frequency – MHz
0.81
Figure 14. Transfer Characteristic of the Transmit Filters With HP Filter Bypassed
(Complies with ITU G.992.2 PSD requirement)
RECEIVE CHANNEL RESPONSE WITH DIFFERENT EQUALIZER SETTINGS
20
0
–20
–40
–60
–80
Gain – dB
–100
–120
–140
34
–460
–180
00.20.40.60.81
f – Frequency – MHz
1.21.41.6
Figure 15. Transfer Characteristic of the Receive Filters (including out of band 0–1.6 MHz)
(Complies with ITU G.992.2 PSD requirement)
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
RECEIVE CHANNEL RESPONSE WITH DIFFERENT EQUALIZER SETTINGS
20
10
0
–10
–20
Gain – dB
–30
–40
–50
–60
00.20.30.4
0.1
f – Frequency – MHz
0.50.6
Figure 16. Transfer Characteristic of the Receive Filters (in-band 0–0.6 MHz)
(Complies with ITU G.992.2 PSD requirement)
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
35
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
APPLICATION INFORMATION
Output
Signal
Input signal
need to have
AVDD_RX/2
common mode
voltage
5 TXOUP
6 TXOUM
70 RXINP
71 RXINM
75 HPF2OUTM
72 HPF2OUTP
73 HPF2INM
74 HPF2INP
76 HPF1OUTP
79 HPF1OUTM
77 HPF1INM
78 HPF1INP
To
VCXO
VCXOCNTL 13
External
clock
RESET 38
PLLSEL 42
PWRDN 17
Digital Interface
FSX 22
DGPO 35
GPIO0-GPIO7 23-30
MCLKIN/PLLCLKIN 37
TLFD500PN
to DSP
FSR 21
SDR 20
DVDD 31, 39
DVDD_RX 51
SDX 18
SCLK 19
DVDD_IO 18
DVSS 34,40,41
DVSS_IO 33
DVSS_RX 52,54,57
AVDD_REF 47
AVDD_RX 48
AVDD_RX 68
AVDD_TX 7
AVSS_REF 44
AVSS_RX 49
AVSS_RX 69
AVSS_TX 8
Digital
Power
Supply
Digital
Ground
Analog
Power
Supply
VSS 80
16 COMPDAC1
1.0
1.0
µF
µF
Analog
Power
Supply
15 COMPDAC2
10
µF
50 V MID_RX
0.1
10
µF
µF
14 TXBANDGAP
10
0.1
µF
µF
Figure 17. Typical Application Circuit
43 RXBANDGAP
0.1
µF
Analog
Ground
10
µF
45 REFM
0.1
µF
10
µF
Analog
Ground
46 REFP
0.1
µF
36
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
TLFD500PN
3.3 V INTEGRATED G.LITE ANALOG FRONT END
SLAS207A – JUNE 1999 – REVISED NOVEMBER 1999
MECHANICAL DATA
PN (S-PQFP-G80) PLASTIC QUAD FLATPACK
80
61
1,45
1,35
0,50
60
1
9,50 TYP
12,20
SQ
11,80
14,20
SQ
13,80
0,27
0,17
20
41
0,08
M
40
21
0,05 MIN
0,13 NOM
Gage Plane
0,25
0°–7°
0,75
0,45
1,60 MAX
NOTES: A. All linear dimensions are in millimeters.
B. This drawing is subject to change without notice.
C. Falls within JEDEC MS-026
POST OFFICE BOX 655303 • DALLAS, TEXAS 75265
Seating Plane
0,08
4040135 /B 11/96
37
IMPORTANT NOTICE
T exas Instruments and its subsidiaries (TI) reserve the right to make changes to their products or to discontinue
any product or service without notice, and advise customers to obtain the latest version of relevant information
to verify, before placing orders, that information being relied on is current and complete. All products are sold
subject to the terms and conditions of sale supplied at the time of order acknowledgement, including those
pertaining to warranty, patent infringement, and limitation of liability.
TI warrants performance of its semiconductor products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are utilized to the extent
TI deems necessary to support this warranty . Specific testing of all parameters of each device is not necessarily
performed, except those mandated by government requirements.
CERTAIN APPLICA TIONS USING SEMICONDUCT OR PRODUCTS MAY INVOLVE POTENTIAL RISKS OF
DEATH, PERSONAL INJURY, OR SEVERE PROPERTY OR ENVIRONMENTAL DAMAGE (“CRITICAL
APPLICATIONS”). TI SEMICONDUCTOR PRODUCTS ARE NOT DESIGNED, AUTHORIZED, OR
WARRANTED TO BE SUITABLE FOR USE IN LIFE-SUPPORT DEVICES OR SYSTEMS OR OTHER
CRITICAL APPLICA TIONS. INCLUSION OF TI PRODUCTS IN SUCH APPLICATIONS IS UNDERST OOD TO
BE FULLY AT THE CUSTOMER’S RISK.
In order to minimize risks associated with the customer’s applications, adequate design and operating
safeguards must be provided by the customer to minimize inherent or procedural hazards.
TI assumes no liability for applications assistance or customer product design. TI does not warrant or represent
that any license, either express or implied, is granted under any patent right, copyright, mask work right, or other
intellectual property right of TI covering or relating to any combination, machine, or process in which such
semiconductor products or services might be or are used. TI’s publication of information regarding any third
party’s products or services does not constitute TI’s approval, warranty or endorsement thereof.
Copyright 1999, Texas Instruments Incorporated
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.