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
Loading...
+ 26 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.