TI GC5318

GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
1 Introduction
1.1 Features
•
•
•
•
•
Optimized for CDMA2000-1X and UMTS
Up to 12 UMTS or 24 CDMA2000 Upconverter
Channels
Mixed CDMA2000-1X and UMTS Operation
DUC Output Rates to 125 MSPS
Any DUC Can Sum into Any of Four Output
Ports
•
•
•
•
•
Real/Complex DUC Outputs
Tx Filtering: 6-Stage CIC, 47-Tap CFIR, 63-Tap
PFIR
115-dB SFDR
18-Bit DUC Outputs
1.5-V Core, 3.3-V I/O
1.1.1 Description
The GC5318 is a high-density multi-channel communications signal processor integrated circuit that
provides digital upconversion optimized for cellular base transceiver systems. The device supports both
UMTS and CDMA2000 (CDMA) air interface cellular standards.
The chip provides up to 24 CDMA digital upconverter (DUC) channels or 12 UMTS DUC channels. The
GC5318 can also support a combination of CDMA and UMTS channels. The DUC channels operate
simultaneously.
The chip is ideal for cellular base transceiver systems where a large number of digital radio channels are
required. Each of the 24 CDMA (or 12 UMTS) channels can operate independently.
There are four 18-bit output ports. Each output port can sum any of the DUC channels in a daisy-chain
fashion. This configuration permits creating a stack of CDMA or UMTS signals. These ports can output
either real or complex data. Real output data would generally drive one or more digital-to-analog (D/A)
converters and output the stack of signals at an intermediate frequency (IF). Complex data (at baseband
or an IF) is used when a quadrature modulator upconversion scheme is employed. Complex output data
can also be used when the output stack is further processed using crest factor reduction or power
amplifier predistortion techniques.
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 document.
All trademarks are the property of their respective owners.
PRODUCTION DATA information is current as of publication date.
Products conform to specifications per the terms of the Texas
Instruments standard warranty. Production processing does not
necessarily include testing of all parameters.
Copyright © 2006, Texas Instruments Incorporated
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
1.1.2 Functional Block Diagram
txout_d
txout_c
txout_b
txout_a
tx_lflag
tx_clk_out
Transmit Output Data
(to D/A Converters)
Transmit Output
Interface
DUC0
2CDMA2000-1X
or 1 UMTS
Serial Baseband Transmit Channel Inputs
I&Q
sync
DUC1
2CDMA2000-1X
or 1 UMTS
DUCs 2-9
DUC10
2CDMA2000-1X
or 1 UMTS
sync
DUC11
2CDMA2000-1X
or 1 UMTS
txclk
sync
tx_sync a-d
tdi
Control & Sync
4
tx_sync_out
reset_n
trst_n
tck
control
JTAG
tdo
interrupt
16
tms
d0 - d15
6
a0 - a5
rd_n
wr_n
ce_n
1.1.3 Package/Ordering Information
2
PRODUCT
PACKAGE-LEAD
GC5318
Thermally-Enhanced
Plastic BGA
with Heat Slug – 388
Introduction
PACKAGE
DESIGNATOR
SPECIFIED
TEMPERATURE
RANGE
PACKAGE
MARKING
ORDERING
NUMBER
TRANSPORT
MEDIA,
QUANTITY
ZED
–40°C to +85°C
GC5318IZED
GC5318IZED
Tray, 40
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Contents
1
Introduction ............................................... 1
1.1
Features .............................................. 1
4.1
5
1.1.1 Description ........................................... 1
1.1.2 Functional Block Diagram ............................ 2
1.1.3 Package/Ordering Information ....................... 2
2
GC5318 Operation ....................................... 4
....................... 5
2.2
Sum Chain Shifting and Rounding .................. 21
2.3
GC5318 Output Interface ........................... 24
GC5318 General Control .............................. 25
3.1
Control Data, Address, and Strobes ................ 26
3.2
MPU Timing Diagrams .............................. 26
3.3
Interrupt Handling ................................... 27
3.4
Sync Signals ........................................ 27
3.5
Initialization .......................................... 28
3.6
GC5318 Board Diagnostics ......................... 28
GC5318 Programming ................................. 28
2.1
3
4
Digital Upconvert Block (DUC)
6
cmd5318 Keywords ................................. 29
GC5318 Pin Description............................... 37
..............................................
.............................
5.3
JTAG Signals ........................................
5.4
Factory Test and No Connect Signals ..............
5.5
Power and Ground Signals..........................
5.6
Power Monitoring ....................................
5.7
JTAG ................................................
Specifications ...........................................
6.1
Absolute Maximum Ratings .........................
6.2
Recommended Operating Conditions ...............
6.3
Thermal Characteristics .............................
6.4
Power Consumption .................................
6.5
DC Operating Conditions............................
6.6
AC Characteristics ..................................
5.1
Signals
5.2
Microprocessor Signals
Contents
37
39
40
40
40
40
41
41
41
41
42
42
42
43
3
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
tc_clk_out
tx_flag
txout_d
txout_c
txout_a
Transmit
Output Data
(to D/A Converter)
txout_b
2 GC5318 Operation
Transmit
Output
Interface
DUC 0 Inputs:
Sum Chain Out
For UMTS
For CDMA
I UMTS
I/Q CDMA Ch A
txin_0_a
Q UMTS
I/Q CDMA Ch B
txin_0_b
Frame sync for DUC 0 & 1
DUC 0
2 CDMA2000
or 1 UMTS
Sum Chain
tx_sync_out_0
I UMTS
I/Q CDMA Ch A
txin_1_a
Q UMTS
I/Q CDMA Ch B
txin_1_a
DUC 1
2 CDMA2000
or 1 UMTS
DUC 1 Inputs:
DUCs 2-9
Sum Chain
I UMTS
I/Q CDMA Ch A
txin_10_a
Q UMTS
I/Q CDMA Ch B
txin_10_b
DUC 10 Inputs:
Frame sync for DUC 10 & 11
DUC 11 Inputs:
(baseband data)
DUC 10
2 CDMA2000
or 1 UMTS
tx_sync_out_5
I UMTS
I/Q CDMA Ch A
txin_11_a
Q UMTS
I/Q CDMA Ch B
txin_11_b
Sum Chain
DUC 11
2 CDMA2000
or 1 UMTS
txclk
tx_synca
Transmit
Input Syncs
tx_syncb
tx_syncc
Transmit Syncs
tx_sync_out
tx_syncd
Figure 2-1. GC5318 Block Diagram
The GC5318 provides up to 24 CDMA2000 or 12 UMTS digital upconversion (DUC) channels. There are
12 DUC blocks, DUC 0 through DUC 11, as shown in the block diagram of Figure 2-1. Each block can be
configured as a single UMTS channel or as two CDMA channels.
4
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
The outputs of all DUCs drive four independent complex sum chains. Any DUC can contribute (or not) to
any or all of the four sum chains. The output of a DUC block sum chain drives the sum chain input of the
next block. The first DUC to output data is DUC0, while the last is DUC11. The two or four outputs of a
DUC are the sum of all the contributing channels of all the higher numbered DUC blocks and itself. The
sum chain inputs of DUC 11 are grounded. Within the chain, all DUC blocks from 0 up to the highest
numbered DUC in use must be turned on; otherwise, the sum chain is broken.
The transmit output interface takes the four summed chains of DUC output data from the output of DUC 0,
then scales and rounds them to a user-programmed number of bits. Composite power meters with
programmable integration periods and intervals compute the power in each of the four output streams.
The data is then formatted for output over the four tx_data_out ports.
2.1 Digital Upconvert Block (DUC)
Pilot
Insertion
(UMTS Only)
4 Sign on Top
16 Data
15 Zeros on Bottom
16
Serial
I, Q
16
34
Serial
Interface
5 Guard on Top
18 Data
13 on Bottom
36
18
Round
Gain
Frequency
RMS
Power Meter
Phase
Programmable
FIR Filter
Interp by 2
18
Compensating
FIR Filter
Interp by 2
18
Six Stage
CIC Filter
Interp 4-32
18
32
NCO
16
18
Delay Adjust
CDMA A & B Channels, or 1 UMTS Channel
From Previous Transmit Channel
Sum1
Input
Sum2
Input
Sum2
Output
Sum4
Input
Enable
Enable
Enable
Enable
Sum1
Output
Sum3
Input
Sum3
Output
Sum4
Output
To Next Transmit Channel
Figure 2-2. Digital Upconvert Block
GC5318 Operation
5
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
This section describes the functions available in each of the 12 DUC blocks. Each block has its own
register set and may be programmed individually. The final output rates of all DUC blocks must match
since they are added together.
Figure 2-2 illustrates the different signal processing blocks and general signal processing flow of an
individual transmit channel. Within a block, a single set of hardware performs these functions for one
UMTS signal or two CDMA signals. When processing two CDMA signals, the gain, round, power meter,
PFIR and CFIR blocks are time-shared to process both signals with one set of hardware. Each DUC can
support one UMTS channel or two CDMA channels.
Each block accepts baseband serial data. At this point, the gain can be adjusted and a pilot sequence can
be summed with the data. Baseband data gain adjusted and the pilot signal power can be measured. The
programmable FIR filter (PFIR) is used to pulse shape the data and interpolates by a factor of two. The
compensating CIC filter (CFIR) compensates for the roll-off of the following CIC filter and also interpolates
by a factor of two. The CIC filter performs additional interpolation; this additional interpolation is
programmable. The delay adjust block permits the channel's delay to be adjusted relative to all other DUC
channels.
The interpolated, filtered, and delayed data is then tuned to a user-programmed frequency with a digital
mixer and oscillator. The DUC output data then drives four independent sum chain paths, where output
data from each DUC can be summed into four composite streams.
Each function block is described in greater detail in subsequent sections.
Table 2-1. Programming
VARIABLE
DESCRIPTION
cdma_mode
2.1.1
When set, the DUC block is in CDMA2000 mode.
Transmit Serial Input Interface
1 UMTS
2 CDMA
1 UMTS
I
CH A
Interleaved I, Q
txin_X_a
IMSB
IMSB-1
CH B
Interleaved I, Q
txin_X_b
Q
CDMA DUC Channel A
CDMA DUC Channel B
QMSB
2
Frame Strobe
Pins From Adjacent DUC
Or 1 UMTS Channel
DUC Block
Frame
Delay
4
Clkdiv
Sync
Shared between two DUC blocks, that is: 2k and 2k + 1, k = 0 to 5.
QMSB-1
Figure 2-3. Transmit Serial Input Interface
6
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Each DUC block has two serial input data pins; see Figure 2-3. These pins are used to transfer I/Q
baseband data into the DUC channel for interpolation, filtering, and tuning to a carrier frequency. The
channel configuration of the block determines how these pins are used.
When the block is configured for two CDMA channels, one pin (txin_X_a) accepts serial data for signal A;
the other pin (txin_X_b) for signal B. Input I and Q data, programmable up to 18 bits, are multiplexed over
the serial input pin starting with the most significant I bit. The maximum input bit rate is txclk. The interface
can be programmed to accept up to 32 bits, but only the upper 18 bits will be used as input signal data.
The most significant bit is sent first.
When the block is configured for a single UMTS channel, the txin_a is for I data, and txin_b carries the Q
data.
The four-pin mode is less common. It employs another two pins from the adjacent (2k + 1) DUC,
sacrificing the use of that DUC ito allow reduced data rate on the serial pins. The I data (Imsb, Imsb-1) are
carried on txin _(2k) _a and txin _(2k)_b, while the Q data (Qmsb Qmsb-1) are carried on txin _(2k + 1) _a
and txin _(2k + 1) _b.
Each pair of blocks (2k and 2k + 1) shares the clock division, frame delay, sync generation, and a frame
strobe output pin.
A programmable clock divider circuit can be used to specify the serial bit rate with respect to txclk. The
divider is programmed as txclk / (1+serp_trans_clkdiv). The clock divider circuit is synchronized using a
general sync block discussed in another section of this document.
The frame sync interval can be programmed from 1 to 127 bit-periods (which are divided clocks).
The number of bits in a word is set as (serp_tran_bits+1).
The frame strobe is an output from the GC5318 that indicates when the msb is expected. The frame
strobe can be programmed to arrive from 0 to 3 bit clocks ahead of when the msb is expected via the
serp_tran_fsdel parameter. The source must transmit all of its data before the next frame strobe is
generated. Use of the frame strobe is optional. The sync (ssel_serial) parameter determines when the
msb is expected.
The parameter chosen must satisfy the following constraints:
• serp_tran_fsinv × (serp_tran_clkdiv+1) = 4 × (cic_interp_decim+1)
• serp_tran_fsinv ≥ (serp_tran_bits+1)2 for CDMA mode
• serp_tran_fsinv ≥ (serp_tran_bits+1) for UMTS mode
• serp_tran_fsinv ≥ (serp_tran_bits+1)0.5 for four-pin mode
NOTE
For half-rate data (when serp_tran_clkdiv = 1), the MSB of the input data stream is
captured on the fourth rising edge of txclk, after txsync occurs. For full-rate data (when
serp_tran_clkdiv = 0), the MSB of the input data stream is captured on the third rising
edge of txclk, after txsync occurs.
See Figure 2-4 for a diagram of the serial transmit input timing.
GC5318 Operation
7
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
tpd
tSU
tH
tSU
tH
txclk
txsync
tx_sync_out
2txclk
(programmable)
2txclk
txin
(serial input data)
MSB
MSB-1
serp_tran_fsdel = 0
serp_tran_clkdiv = 1
Figure 2-4. Serial Input Timing
Table 2-2. Programming
VARIABLE
serp_tran_bits(4:0)
DESCRIPTION
Number of serial input bits in a word – 1. That is, 10001 = 18 bits
serp_tran_fsinvl(6:0)
Frame sync interval in bits
serp_tran_fsdel(1:0)
The number of serial bits after frame strobe that the data MSB is expected.
serp_tran_4pin
serp_tran_clkdiv(3:0)
ssel_serial(2:0)
0 = 2 pin input mode. Applies to UMTS mode for separate I and q data bits, as well as CDMA mode where one
pin is for interleaved I/Q data for the CDMA A channel and another pin for interleaved I/Q data for the CDMA B
channel.
1 = 4 pin mode. Applies to UMTS mode where the channel has two bits for I data (Imsb and Imsb-1) and two bits
for Q data (Qmsb and Qmsb-1)
Serial input data bit clock divider factor – 1
Sync source
The parameters are set for a pair of DUC blocks; that is, for 2k and 2k + 1 DUCs, where k = 0 to 5.
2.1.2
Gain
Figure 2-5. Gain Block
The transmit gain block, shown in Figure 2-5, is a multiplier that increases or decreases the level of the
input data. The unsigned 16-bit gain word is interpreted with the binary point three bits down from the
MSB. It multiplies the input data by (gain word/8192). The maximum gain is therefore 65535/8192. There
are different gain registers for the A and B signals in CDMA mode.
8
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
A transfer register in combination with a sync (ssel_gain) synchronizes gain changes across multiple
channels.
Table 2-3. Programming the Gain
VARIABLE
2.1.3
DESCRIPTION
gainfora(15:0)
Gain for the A-side DUC. Interpreted as gainfora/8192; unsigned.
gainforb(15:0)
Gain for the B-side DUC. Interpreted as gainforb/8192; unsigned.
ssel_gain(2:0)
Sync source
UMTS Pilot Code Insertion
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
In
0
Upper Register Load
I State
16
15
14
13
12
11
10
9
8
Qln
7
16
I
Q
In Qln I
0 -G G
0
18
17
Output
Logic
6
5
4
3
2
1
0
Gain
Lower Register Load (always loaded with all 1s)
16
0
1
G
1
0
-G -G
1
1
G
G
-G
16
Q
A
Reg
Negate
Synchronization
16-Bit Counter for
Gain Negation
Sync
Sync Delay Counter
16 Bits
(in samples)
16
Delay
Figure 2-6. Pilot Code Insertion Logic
The pilot code insertion block, shown in Figure 2-6, generates UMTS pilot scrambling sequences and
should be disabled by setting gain to zero when transmitting CDMA. The pilot sequence is summed with
the UMTS input baseband data prior to PFIR filtering.
The pilot sequence is complex and generated from two 18-bit shift registers, each with a unique set of
feedback taps. Specific taps are exclusive-or combined to form the I and Q streams. The streams are then
modified by a user-programmed complex gain value. The gain word G is a signed 16-bit value. The output
sequence is ±G. Setting gain to zero turns off pilot insertion.
NOTE
Gain MUST be set to zero for CDMA operation.
The upper 18-bit shift register is programmed with a starting sequence based on the desired primary
scrambling code (PSC). There are 512 start sequences for all of the base station transceiver (BTS) codes.
The lower register always starts with a string of all 1s.
GC5318 Operation
9
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
When diversity channels are employed, a counter in the synchronization block toggles the
value in a prescribed fashion. The UMTS frame starts with positive gain for 256 chips,
negative gain for 512 chips, then toggles again to positive gain for 512 chips, and so on
the frame. The last 256 chips of the frame are negative gain. This sequence repeats
frames.
sign of the gain
then toggles to
until the end of
for subsequent
Table 2-4. Programming
VARIABLE
DESCRIPTION
pilot_psc(15:0)
Lower 16 bits of the 18-bit pilot LFSR initial sequence.
pilot_psc(17:16)
Upper two bits of the 18-bit pilot LFSR initial sequence.
pilot_diversity
Sets main or diversity pilot generation. 0 = main, 1 = diversity
pilot_delay(15:0)
Unsigned delay value (in chips) from sync event. 0 to 38399 chips.
pilot_gain_0(15:0)
Gain value. pilot_gain_0 and pilot_gain_1 must be set to the same value for proper operation. Must be set to 0 for
CDMA operation.
pilot_gain_1(15:0)
Gain value. pilot_gain_0 and pilot_gain_1 must be set to the same value for proper operation. Must be set to 0 for
CDMA operation.
ssel_pilot(2:0)
2.1.4
Sync source
RMS Power Meter
I
18
37
Q
Top
32
Integrate
50 Bits
Register
32
RMS Power
18
Transfer
Clear
pmeter_sync_delay_duc pmeter_interval_duc
7
Sync
13
Interval
Counter
9 Bits
(in increments
of 64 samples)
Sync Delay
Counter
7 Bits
(in samples)
Power Measurement Time Line
Sync Delay
pmeter_integration_duc
9
Integration Period Counter
13 Bits
(increments in samples)
Interrupt
Integration Time
Interrupt
Integration Time
Interrupt
Interrupt
Integration Time
Time
Interval Time
Sync
Event
Integration
Start
Interval Time
Integration
Start
Integration
Start
Figure 2-7. RMS Power Meter
10
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Each transmit channel includes an RMS power meter used to measure the RMS power within the channel.
The RMS power meter samples the I and Q data stream before the PFIR filter; see Figure 2-7. Both 18-bit
I and Q data are squared, summed, and then integrated over a time determined by
pmeter_integration_duc (13 bits). The integration time = 4 x pmeter_integration_duc + 1 (in units of a
sample period or generally a chip period).
A programmable 9-bit interval counter sets the interval over which power measurements are repeated.
The timer counts in increments of 64 samples. The interval time = 64 (pmeter_interval_duc+1). The
interval time must be greater than (not equal to) the integration time.
The power measurement process starts with a sync event (ssel_pmeter). The integration starts at sync
event + 3 chips + sync_delay. The 7-bit delay register permits delays from three to 130 samples after
sync. The integration continues until the integration count is met. At that point, the top 32 bits of the 50-bit
accumulator are transferred to the read register and an interrupt is generated that indicates the power
value is ready to read. The interval counter continues until the programmed interval count is reached.
When reached, the integration counter and the interval counter start over again. Each time the integration
count is reached the upper 32 bits are again transferred to the read register, overwriting the previous
value and sending an interrupt signifying the power value can be read. Failure to read the data before the
next interrupt signal results in overwriting the previous interval measurement.
Sync ssel_pmeter starts the process. Whenever a sync is received, all counters are reset to zero no
matter what the status may be.
For UMTS, I and Q are calculated and the integrated power is read. In CDMA mode the power is
calculated for both the A and B signals, producing two 32-bit results.
For CDMA mode, the integration time is slightly longer. The power read in CDMA mode with a dc input is:
• A power: [ I2 × (X × 4 + 1) + Q2 × (X × 4 + 0) ] × 2-18. Note: One Q sample is missing from the
integration.
• B power: [ I2 × (X × 4 + 1) + Q2 × (X × 4 + 1) ] × 2-18
where X is the integration count.
Table 2-5. Programming
VARIABLE
DESCRIPTION
pmeter_result_a_lsb(15:0)
Lower 16 bits of the A channel power measurement.
pmeter_result_a_msb (31:16)
Upper 16 bits of the A channel power measurement.
pmeter_result_b_lsb (15:0)
Lower 16 bits of the B channel power measurement result. Only available in CDMA mode.
pmeter_integration_duc(12:0)
Integration time = 4 x pmeter_integration_duc + 1.
pmeter_sync_delay_duc(6:0)
Sync delay count in samples.
pmeter_interval_duc(9:0)
ssel_pmeter(2:0)
pmeter_sync_disable
Interval time = 64 (pmeter_interval_duc + 1). Interval time must be greater than (not equal to) integration
time.
Sync source options.
Turns off sync to the channel's power meter
GC5318 Operation
11
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
2.1.5
Filter Chain
GC5318 transmit filtering is performed in three stages:
• Interpolate by two pulse-shape filtering using the programmable FIR filter (PFIR)
• Interpolate by two compensation filtering using the programmable compensating FIR filter (CFIR)
• High-rate interpolation (4 to 32) using the six stage cascade-integrate comb filter (CIC)
Figure 2-8. DUC Filter Chain
The purpose of the transmit filter chain, shown in Figure 2-8, is to interpolate the input signal data up to
the mixer clock rate, nominally 122.88 MHz. The following table provides two examples of how the
interpolation can be allocated among the three different filters for both CDMA and UMTS.
Table 2-6. Example UMTS and CDMA2000 DUC Transmit Modes
Input Rate Rate
PFIR Interpolation
CFIR Interpolation
CIC Interpolation
Overall Interpolation
CDMA
1.2288 MSPS
2
2
25
100
UMTS
3.84 MSPS
2
2
8
32
2.1.5.1 Programmable FIR Filter
The programmable FIR filter (PFIR) pulse shapes the baseband signal data and interpolates by a fixed
factor of two.
The PFIR length is programmable. This permits turning off taps and saving power if short filters are
appropriate. The maximum PFIR filter length is a function of GC5318 clock rate and input sample rate and
is limited by the number of coefficient memory registers. The number of taps available in CDMA mode
ranges from 31 to 63. In UMTS mode it ranges from 15 to 63. In both cases, the number of taps is one
less than a multiple of four (31, 35, ... 59, 63).
Subject to the above range, the maximum number of taps available in each mode is:
• UMTS Mode: 2 × ( txclk × input sample rate)
• CDMA Mode: txclk × input sample rate
Assuming a txclk of 122.88 MHz, both UMTS (3.84 MSPS) and CDMA (1.2288 MSPS) modes provide 63
taps. The same PFIR coefficients are used for both the A and B channels in CDMA mode.
The PFIR filter consists of 32 forward and reverse data RAM cells each 36 bits in width. The coefficient
memory provides storage for up to 63 unique 18-bit taps. A 19 x 18 multiplier and full-precision
accumulator form the filter convolution. An optional (pfir_gain) up-shift of one follows. Finally, the output is
hard-limited.
The PFIR gain is:
Gain = sum(coefficients) × 2^[cfir_gain-19]
12
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-7. Programming
VARIABLE
crastsrttap_pfir(4:0)
cdma_mode
symmetric_pfir
DESCRIPTION
Number of DUC PFIR filter taps is (2 x crastarttap) + 1
When set, puts the CFIR and PFIR blocks in CDMA2000 mode.
Set to '1' if filter is symmetric. This value saves a modest amount of power.
The 18-bit PFIR filter coefficients are loaded by the software cmd5318. The user must provide a coefficient file
with one integer coefficient per line. Note that the PFIR filter coefficients are shared by the A and B signals in
CDMA mode.
2.1.5.2 Compensating FIR Filter
The CFIR filter interpolates by a fixed factor of two and is usually programmed to compensate for the CIC
filter roll-off.
The CFIR filter length is programmable. This programmable length permits turning off taps and saving
power if short filters are appropriate. The maximum CFIR filter length is a function of GC5318 clock rate
and input sample rate and is limited by the number of coefficient memory registers. The number of CFIR
taps in CDMA mode is 31 to 47, while in UMTS mode it is 15 to 31. In both cases, the number of taps is
one less than a multiple of four (31, 35, ... 59, 63).
Subject to the above minimum, maximum, and increment values, the maximum number of taps available
is:
• UMTS Mode: txclk × input sample rate
• CDMA Mode: 0.5 x (txclk × input sample rate)
For example, a txclk of 122.88 MHz in UMTS mode (at 3.84 MSPS) provides 31 taps, while CDMA
(1.2288 MSPS) mode provides 47 taps.
The CFIR coefficients are shared by the A and B signals in CDMA mode.
CFIR gain is:
Gain = sum(coefficients) x 2cfir_gain-19
Table 2-8. Programming
VARIABLE
crastsrttap_cfir(4:0)
symmetric_cfir
cfir_gain
DESCRIPTION
These bits define the number of taps that CFIR uses for the filtering. DUC CFIR: (2 x crastarttap_cfir) +1.
Note: crastarttap_cfir must be odd.
Set to '1' if filter is symmetric. Saves a bit of power.
CFIR gain adjustment.
The 18-bit PFIR filter coefficients are loaded by the software cmd5318. The user must provide a coefficient file with
one integer coefficient per line. Note that the PFIR filter coefficients are shared by the A and B signals in CDMA
mode.
GC5318 Operation
13
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
2.1.5.3 CIC Filter
Force 0
-m1
-m2
Z
In
18
-m3
Z
19
-m4
Z
20
-m5
Z
21
-m6
Z
22
N
Z
23
24
Zero Pad
by 3-31
24
m1, m2, m3, m4, m5, m6 = 1 or 2
-1
Z
-1
Z
29
-1
Z
34
-1
Z
39
-1
Z
44
-1
Z
50
Shift
Shift
0-31
and
Round
18
Out
Figure 2-9. CIC Filter
The six stage CIC filter (as shown in Figure 2-9) interpolates over a programmable range from four to 32.
The filter is made up of six banks of 24-bit subtractor sections followed by six banks of integrator sections.
Each of the six subtractor sections can be independently programmed with a differential delay of one or
two. A shift block follows the last integration stage and can shift the 50-bit accumulated data down by
31-TCIC_SHIFT bits yielding 18-bit output data.
The CIC filter exhibits a droop across its frequency response. Usually the preceding CFIR filter
precompensates for the CIC droop with a gradually rising frequency response. However, it is also possible
to provide the precompensation in the PFIR filter.
CIC interpolation filters can become unstable if an external event (such as a cosmic particle) disturbs a
storage node in the CIC integrator section. This instability can add a bias that subsequently integrates out
of control. The GC5318 CIC employs a patented method to first detect this bias, then automatically flush
and reset the filter. Register bits are available to disable and to test this auto-flush feature. A maskable
interrupt becomes active if a CIC error occurred.
The gain of the CIC filter is:
Ncic5 x 2(number of stages where M=2) x 2(CIC_SCALE - 31)
where CIC_SCALE is 0 to 31. The CIC gain should have a value <1.0. Ncic is the interpolation ratio and is
programmed as cic_interp_decim + 1.
Since the CIC output is full rate for both UMTS and CDMA, a complete hardware path is required for each
of the signals A and B from this point on in the transmit signal path. For CDMA, there are four
independent CIC filters (I/Q for signal A and I/Q for signal B). For UMTS, the two signal B CIC filters are
disabled.
14
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-9. Programming
VARIABLE
DESCRIPTION
cic_interp_decim(4:0)
The CIC interpolation is Ncic = cic_interp_decim + 1. This ratio applies to both A and B channels of the DUC
block in CDMA mode. Legal values for cic_interp_decim are 3 to 31.
cic_scale_a(4:0)
The shift value for the A channel. A value of 0 is no shift; each increment in value increases the amplitude of
the shifter output by a factor of 2.
cic_scale_b(4:0)
The shift value for the B channel. A value of 0 is no shift; each increment in value increases the amplitude of
the shifter output by a factor of 2.
cic_m2_ena_a(5:0)
Sets the differential delay value M for each of the CIC subtractor stages for the A channel. Cic_m2_en_a(0)
controls m1, cic_m2_en_a(5) controls the m5. A set bit programs the differential delay M to 2; if cleared, M is
programmed to 1.
cic_m2_ena_b(5:0)
Sets the differential delay value M for each of the CIC subtractor stages for the B channel.
ssel_cic(2:0)
Sync source
cic_auto_flush_dis(3:0)
When set, disables the CIC auto-flush. Bits {0, 1, 2, 3} correspond to CICs for {CDMA-A I data, CDMA-A Q
data, CDMA-B I data, CDMA-B Q data} sections.
cic_auto_flush_test(3:0)
On rising, forces a CIC overflow error. Program to 0, then to 1 for edge to occur. Bits {0, 1, 2, 3} correspond
to CICs for {CDMA-A I data, CDMA-A Q data, CDMA-B I data, CDMA-B Q data} sections.
cic_auto_flush_clear(3:0)
On rising, clears a CIC overflow condition. Program to 0, then to 1 for edge to occur. Bits {0, 1, 2, 3}
correspond to CICs for {CDMA-A I data, CDMA-A Q data, CDMA-B I data, CDMA-B Q data} sections.
2.1.6
Adjustable Channel Delay
Figure 2-10. Delay Adjustment
The channel delay adjust function, shown in Figure 2-10, permits the user to add a programmable time
delay in each of the upconverter paths. This function is used to calibrate time delays in multiple channels
in the overall base transceiver system. The adjustable delay compensates for analog elements external to
the digital upconversion such as cables, splitters, analog upconverters, filters, etc., and for differential
delay between channels within the GC5318. There is an additional delay of two output sample times for
each pair of DUC blocks to allow for pipelining of the sumchain (specifically, DUC0 and 1 have the same
delay; DUCs 2 and 3 are the same but are two output sample times larger than DUC0 and 1, etc.).
There are two elements that need to be considered with respect to programming the delay: the decimation
and delay memory blocks.
The decimation function reduces the sample rate from txclk to the desired output rate. Output decimation
is required for CDMA mode, and is also used for Interleaved IQ outputs. The decimation amount is set by
parameter (tadj_interp_decim+1). Phasing of the decimation operation permits finer delay resolution. The
3-bit delay offset parameter permits finer delay resolution in steps of the reciprocal of the GC5318 tx clock
rate. At 122.88 MHz, this value would equate to a time delay resolution of 8.1 ns (1/32 chip for UMTS,
1/100 of a chip for CDMA). The offset may be set from 0 to tadj_interp_decim.
The coarse delay adjustment is done using a delay memory of 64 memory locations by 36 bits (18 for I
and 18 for Q). Read and write pointers in the memory are separated by tadj_offset_coarse. Data written
into a location is read out tadj_offset_coarse output sample times later. 24 locations are needed to
equalize the time delay within the GC5318 for various channels. The remaining 40 locations provide a
total delay of up to about 1.3 µs when the DUC output data rate is 30.72 MSPS.
A sync signal permits the decimation operation to be synchronized over multiple channels.
GC5318 Operation
15
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-10. Programming
VARIABLE
DESCRIPTION
tadj_offset_coarse_a(5:0)
Read offset into the 64-element memory for the A channel DUC. When tadj_offset_coarse_a = 62, then the
delay is –2. When tadj_offset_coarse_a = 63, then the delay is –1. For all other values, the resulting delay is
equal to the value.
tadj_offset_coarse_b(5:0)
Read offset into the 64-element memory for the B channel DUC when in CDMA mode. Note: When
tadj_offset_coarse_b = 62, then the delay is –2. When tadj_offset_coarse_b = 63, then the delay is –1. For all
other values, the resulting delay is equal to the value.
tadj_offset_fine_a(2:0)
Controls the zero offset (fine adjust) for the A side of the DUC. The fine adjust is mapped differently.
tadj_offset_fine_b(2:0)
Controls the zero offset (fine adjust) for the B side of the DUC when in CDMA mode. The fine adjust is
mapped differently.
tadj_interp_decim(2:0)
The decimation value (1, 2, 4, or 8) for the DUC. Same for A and B channels when in CDMA mode.
ssel_tadj_fine(2:0)
ssel_tadj_coarse(2:0)
Selects the sync source for the fine time adjust.
Selects the sync source for the coarse time delay adjust.
For a decimation value of 1, the only legal setting is 0; there is no fine adjustment since there is no
decimation moment.
tadj_offset_fine = 0 → fine delay by 0
For a decimation value of 2:
tadj_offset_fine = 0 → fine delay by 0
tadj_offset_fine = 1 → fine delay by 1
16
For a decimation value of 4:
tadj_offset_fine = 0 → fine
tadj_offset_fine = 1 → fine
tadj_offset_fine = 2 → fine
tadj_offset_fine = 3 → fine
delay
delay
delay
delay
by 0
by –1
by –2
by 1
For a decimation value of 8:
tadj_offset_fine = 0 → fine
tadj_offset_fine = 1 → fine
tadj_offset_fine = 2 → fine
tadj_offset_fine = 3 → fine
tadj_offset_fine = 4 → fine
tadj_offset_fine = 5 → fine
tadj_offset_fine = 6 → fine
tadj_offset_fine = 7 → fine
delay
delay
delay
delay
delay
delay
delay
delay
by 0
by –1
by –2
by –3
by –4
by –5
by –6
by 1
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
2.1.7
Mixer
From Channel Delay
I
Q
18
21
To Sum Chain
Round
18
21
20
COS
20
SIN
From NCO
Figure 2-11. Mixer
The mixer is a complex multiplier which takes the baseband I and Q data that has been previously
pulse-shaped and interpolated, and translates to a carrier frequency programmed into the NCO. The mixer
data size is 18 bits for the signal path and 20 bits for the NCO path, as shown in Figure 2-11.
The gain through the mixer is –12 dB. It can be increased by 6 dB through a control bit. It is
recommended that this extra gain always be used. The output is then rounded to 21 bits.
Mixer gain = 2mixer_gain - 2
The mixer output of each channel is combined in daisy-chain fashion in four sum chain adder blocks that
are described in a subsequent section.
For CDMA, the maximum output rate is txclk/2. The maximum output rate for UMTS is txclk.
Table 2-11. Programming
DESCRIPTION
mixer_gain
When asserted, adds 6 dB of gain in the mixer. Should always be set.
NCO
32
Frequency Sync
20
Frequency Word
32
Reg
32
23
Reg
18
Aligned to
Bottom 5 Bits
Clr
Aligned to
Top 32 Bits
2.1.8
VARIABLE
Zero Phase Sync
Phase Offset Sync
Reg
SIN/COS
Table
20
COS
SIN
5
Dither
Gen
16
Phase Offset
Dither Sync
Figure 2-12. NCO
The NCO, shown in Figure 2-12, is a digital complex oscillator that translates (or upconverts) interpolated
and filtered baseband signals to a programmable carrier frequency.
The block produces programmable complex digital sinusoids by accumulating a delta phase frequency
word that is programmed by the user. A phase offset can be added to the accumulated value if desired for
channel calibration purposes. The output of the accumulator is a phase value that indexes into a Sin/Cos
ROM table which produces the complex sinusoid.
A 5-bit dither generator is provided, and generates a small level of digital pseudo-noise that is added to
the phase accumulated value. It is useful for reducing NCO spurious outputs.
GC5318 Operation
17
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-12. Programming
VARIABLE
DESCRIPTION
dither_ena
When set, turns dither on. Clearing turns dither off.
The NCO spurious levels are shown in Figure 2-13 and Figure 2-14. Added phase dither randomizes the
ROM lookup slightly, thereby creating the ROM lookup error—spreading the spurious energy around
rather than concentrating it in a few frequencies. The phase dither is added below the LSB of the ROM
lookup. If the tuning frequency has no high bits more than 17 bits below the msb, the phase dither has no
effect. If the tuning frequency is a multiple of Fs/96, then an initial phase offset of four often reduces NCO
spurs. Figure 2-13 and Figure 2-14 show the spur level performance of the NCO without dither, with
dither, and with a phase offset value.
a) Worst Case Spectrum Without Dither
b) Spectrum With Dither (Tuned to Same Frequency)
Figure 2-13. Example NCO Spurs With and Without Dithering
a) Plot Without Dither or Phase Initialization
b) Plot With Dither and Phase Initialization
Figure 2-14. NCO Peak Spur Plot
18
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
The tuning frequency is specified as a 32-bit frequency word and is programmed as two sequential 16-bit
words over the control port. The NCO operates at the same speed as the txclk/(tadj_interp_decim + 1).
The frequency resolution is simply the FCLK/ 232. The NCO frequency resolution is simply the FCLK/ 232. As
an example, at an input clock rate of 61.44 MHz, the frequency step size would be approximately 14 MHz.
The frequency word is determined by the formula:
Frequency word (in decimal) = 232 x Tuning Frequency/FCLK
NOTE
Frequency tuning words can be positive or negative valued. Specifying a positive
frequency value translates baseband frequencies upward. Specifying a negative tuning
frequency translates baseband frequencies downwards.
Table 2-13. Programming
VARIABLE
DESCRIPTION
phase_add_a(31:0)
32-bit tuning frequency word for the A signal when in CDMA mode. Also for UMTS mode.
phase_add_b(31:0)
32-bit tuning frequency word for the B signal when in CDMA mode. Not used in UMTS mode.
The phase of the NCO Sin/Cos output can be adjusted relative to the phase of other channel NCOs by
specifying a phase offset. The phase offset is programmed as a 16-bit word, yielding a step size of about
5.5 milliDegrees. The phase offset word is determined by the following formula:
Phase Offset Word = 216 x Offset_in_Degrees / 360; or,
Phase Offset Word = 216 x Offset_in_Radians / 2π
Table 2-14. Programming
VARIABLE
DESCRIPTION
phase_offset_a(15:0)
16-bit phase offset word for the A signal when in CDMA mode. Also for UMTS mode.
phase_offset_b(15:0)
16-bit phase offset word for the B signal when in CDMA mode. Not used in UMTS mode.
Various synchronization signals are available that are used to synchronize the NCOs of all channels with
respect to each other. Frequency sync and phase offset sync determine when frequency and phase offset
changes occur. For example, generating a frequency sync after programming the two frequency words
causes the NCO (or multiple NCOs) to change frequency at that time, rather than after each of the three
frequency words is programmed over the control bus. The NCO accumulator reset sync signal is used to
force the sine and cosine oscillators to their zero phase state. Dither sync can be used to synchronize the
dither generators of multiple NCOs.
Table 2-15. Programming
VARIABLE
ssel_nco(2:0)
ssel_dither(2:0)
ssel_freq(2:0)
ssel_phase(2:0)
DESCRIPTION
Sync source for NCO accumulator reset
Sync source for NCO dither reset
Sync source for NCO frequency register loading
Sync source for NCO phase register loading
GC5318 Operation
19
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
2.1.9
Sum Chain
Enable
Enable
Enable
Enable
Enable
Enable
26
CDM A
DUC B
Or 1 UMTS Channel
DUC 11
CDM A
DUC A
CDM A
DUC B
Or 1 UMTS Channel
DUC 1
SUM 3
Enable
Enable
Enable
Enable
Enable
DUCs
2-10
SUM 2
Enable
Enable
Enable
Enable
Enable
Enable
Enable
26
CDM A
DUC A
SUM 1
To Scale and
Round Block
26
CDM A
DUC B
SUM 0
Enable
Enable
Enable
Enable
Enable
Enable
26
CDM A
DUC A
Or 1 UMTS Channel
DUC 0
Figure 2-15. Sum Chain
The sum chain is a daisy-chain of all DUC outputs into four independent composite output streams. Each
DUC output drives four complex adders, each summing the DUC contribution into the sum chain.
The DUC output data driving the adders is 21 bits. The sum chain partial sum outputs are 26 bits wide to
allow for word growth. Each DUC output can contribute to any of the four sum chains via programmable
enable lines. The output of the last daisy-chained sum is then the composite of all of the 24 CDMA or 12
UMTS channels. There must be no DUCs powered down within a sum chain that have lower numbers
than those that are active. This condition would break the chain. For example, if DUCs 3-5 are used and
active, DUCs 0-2 must not be powered down.
As shown in Figure 2-15, each DUC contains a portion of the sumchain. Individual sum chain outputs can
be summed together using the parameters below.
20
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-16. Programming
VARIABLE
DESCRIPTION
sumchn_sel_a(3:0)
Enable bits for signal A contribution to the four sum chain outputs.
Signal A Out
0000 Signal A added to no busses
0001 Signal A I/Q to bus0
0010 Signal A I/Q to bus1
0100 Signal A I/Q to bus2
1000 Signal A I/Q to bus3
Note: Signal A output can contribute to any combination of the four sumchain outputs. The above 4-bit code can
range from 0 to 15.
sumchn_sel_b(3:0)
Enable bits for signal B contribution to the four sum chains (only when in CDMA mode).
Signal B Out
0000 Signal B added to no busses
0001 Signal B I/Q to bus0
0010 Signal B I/Q to bus1
0100 Signal B I/Q to bus2
1000 Signal B I/Q to bus3
Note: Signal B output can contribute to any combination of the four sumchain outputs. The above 4-bit code can
range from 0 to 15.
2.2 Sum Chain Shifting and Rounding
Figure 2-16. Final Sum Chain Scale and Round Block
Summed data is scaled from 26 bits down to 18 bits, as shown in Figure 2-16. The desired 18 bits can be
taken anywhere over the 26-bit sum chain output window via a programmable register. These 18 bits can
range from sumchain(25:8) on the top end to sumchain(17:0) on the bottom end of the 26-bit output.
The scaled data is then saturated to 18, 16, 14, or 12 bits. Rounded data is MSB-justified with the bottom
bits zeroed. For example, 12-bit rounding would output data to the top 12 bits of the 18-bit word and the
bottom 6 bits would be zeroed.
Gain = 2interf_scale-5
GC5318 Operation
21
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-17. Programming
VARIABLE
DESCRIPTION
interf_scale_0(3:0)
interf_scale_1(3:0)
interf_scale_2(3:0)
interf_scale_3(3:0)
Selects the sum chain scale value for each of the four sum chains. The 18-bit output can be slid anywhere across
the 26-bit window.
0000 = sumchain(25:8)
0001 = sumchain(24:7)
…
0111 = sumchain(18:1)
1000 = sumchain(17:0)
interf_round(1:0)
2.2.1
Specifies the rounding of all four sum chains.
00 = 18 bits
01 = 16 bits
10 = 14 bits
11 = 12 bits
Power Meters
Figure 2-17. Power Meter Block
The composite power in each of the four sum chains can be measured using power meters similar to
those used in the individual DUC blocks.
There are four composite RMS power meters, one for each of the four sum chains, as shown in
Figure 2-17. Each of the above power meters is independently programmable with respect to the
measurement period, interval, and delay from sync. The following two sections describe the sum chain
power meters in more detail.
22
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
2.2.1.1 Composite RMS Power Meter
I
18
36
37
From
Sum Chain
Q
18
Register
32
RMS Power
36
Transfer
Clear
pmeterX_sync_delay
pmeterX_interval
9
Sync
Integrate
58 Bits
21
Sync Delay
Counter
9 Bits
(in 8 sample
increments)
21
Interval
Counter
21 Bits
(in samples)
Power Measurement Time Line
Sync Delay
pmeterX_integration
Integration Period Counter
21 Bits
Interrupt
Integration Time
Interrupt
Integration Time
Interrupt
Interrupt
Integration Time
Time
Interval Time
Sync
Event
Integration
Start
Interval Time
Integration
Start
Integration
Start
Figure 2-18. RMS Power Meter
The GC5318 provides four independent output ports, each of which is the sum of a number of individual
carriers (a sum chain). Four composite RMS power meters measure the RMS power of the combined
carriers in each of the four sum chains. These power meters are similar to those used to measure the
RMS power of each individual channel, but have different counter lengths.
The input to the power meter is the scaled and rounded output of a sum chain. Power is calculated by
squaring each 18-bit I and Q sample, summing and then integrating the summed-squared results into a
58-bit accumulator. The integration time is pmeter_integration (maximum of 21 bits) output sample
periods.
There is a programmable 21-bit interval counter that sets the interval over which power measurements are
repeated. The interval time = pmeter_interval + 1. The interval time must be greater than (not equal to) the
integration time. A measurement integration period is started at the beginning of each interval time.
The process begins with a sync event starting the 9-bit delay counter. After the delay count + 2 samples,
the integration interval is started. The power is calculated for each I and Q sample and added to the 58-bit
accumulator. The integration continues until the integration count is met, at which point the upper 32 bits
of the 58-bit integrator are transferred to the read register and an interrupt is generated. A new
measurement period starts at the end of the interval period.
NOTE
Each of the four composite RMS power meter blocks has its own delay sync, interval, and
integration period counters, as well as separate sync source registers.
GC5318 Operation
23
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 2-18. Programming
VARIABLE
DESCRIPTION
comp_pmeterX_result_lsb(15:0)
Lower 16 bits of the composite power measurement for sum chain X. X = 0, 1, 2, or 3
comp_pmeterX_result_msb(31:16)
Upper 16 bits of the composite power measurement for sum chain X. X = 0, 1, 2, or 3
comp_pmeterX_integration_lsb(15:0)
Lower 16 bits of the 21-bit integration period. X = 0, 1, 2, or 3
comp_pmeterX_integration_msb(20:16)
Upper five bits of the 21-bit integration period. X = 0, 1, 2, or 3
comp_pmeterX_sync_delay(8:0)
comp_pmeterX_interval_lsb(15:0)
comp_pmeterX_interval_msb(20:16)
ssel_comp_pmeter_X(2:0)
Power meter delay sync period. X = 0, 1, 2, or 3
Lower 16 bits of the 21-bit measurement interval. X = 0, 1, 2, or 3
Upper five bits of the 21-bit measurement interval. The Interval time must be greater than the
integration time for each of the four composite power meters. X = 0, 1, 2, or 3
Sync source. X = 0, 1, 2, or 3
2.3 GC5318 Output Interface
I
Sum Chain 0
Q
18
18
18
txout_a
I
txout_b
Sum Chain 1
Transmit Output Interface
Q
I
Sum Chain 2
Q
txout_c
I
txout_d
Sum Chain 3
Q
Format
2
txclk_out
tx_i_flag
Figure 2-19. Output Interface
The GC5318 provides four output signal data ports, as shown in Figure 2-19. Each port can be enabled or
disabled. Disabled ports are held low and can also be tri-stated.
Each 18-bit port outputs the sum of the carriers contributing to the composite signal stack. Output data
can be real or complex valued. Complex I/Q data can be output either interleaved over a single output port
or over two ports separately.
Real output data would generally be selected driving a single D/A converter to an IF frequency. Complex
output data would be selected when subsequent post-processing such as power amplifier predistortion is
employed. Complex outputs can also be used to drive a pair of D/A converters (one for I, the other for Q)
for direct I/Q upconversion using a quadrature modulator device.
24
GC5318 Operation
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
I and Q complex output data can be interleaved over a single 18-bit port or simultaneously over two
separate output ports at half the rate. Signal tx_I_flag is active when I data is being output when in
complex output interleaved mode. When complex output data is noninterleaved, I data is output on port 0
and Q data is output on port 1 for sum chain 0. For sum chain 1, I data is output on port 2 and Q data is
output on port 3.
Real output data is output over a single 18-bit port. For CDMA mode, the maximum real output rate is
txclk/2. The maximum real output rate for UMTS mode is txclk.
The maximum complex output rate with I and Q data on separate outputs is txclk/2 for CDMA mode and
txclk for UMTS mode. If the complex output data is interleaved on a single bus, the maximum rate is
txclk/2 for both UMTS and CDMA and the toggle rate between I and Q samples is txclk.
NOTE
The tx_clk_out signal cannot be at full rate (txclk rate) if the mixer is not at full rate (for
CDMA mode, the mixer cannot be at full rate). If a full-rate clock output signal is desired,
the tst_clk signal can be used, with the tst_rate parameter programmed to 0.
Table 2-19. Programming
VARIABLE
interf_ena(3:0)
DESCRIPTION
When bits are set, enables the corresponding outputs. When cleared, outputs are disabled and held low.
interf_real
When set, outputs are real. When cleared, outputs are complex.
interf_interl
When set, complex data is output interleaved.
3-state(3)
When set, turns on tx_data_out3 outputs.
3-state(2)
When set, turns on tx_data_out2 outputs as well as sync_tst, aflag_tst, and clk_tst.
3-state(1)
When set, turns on tx_data_out1 outputs.
3-state(0)
When set, turns on tx_data_out0 outputs as well as tx_iflag and tx_clk_out.
trt_rate
The value here controls the output clock rate on the clk_tst pin. A value of 0 gives a full-rate output clock (txclk rate);
a 1 gives half-rate output clock; a 3 gives 1/4th rate output clock, and so on. The number of txclk cycles for which the
clk_tst signal is high + low = 1 + tst_rate.
3 GC5318 General Control
The GC5318 is configured over a bidirectional 16-bit parallel data microprocessor control port. The control
port permits access to the control registers that configure the chip. The control registers are organized
using a paged-access scheme using six address lines. Half of the 64 addresses (address 32 through
address 63) represent global registers. The other 32 (address 0 through address 31) are paged registers.
This arrangement permits accessing a large number of control registers using relatively few address lines.
Global address 33 is the page register. Writing a 16-bit value to this register sets the page on which future
write or read operations are performed. These paged registers contain the actual parameters that
configure the chip and are accessed by writing/reading address 0 through address 31.
Global registers (address 32 through address 63) are used to read/write GC5318 parameters that are
global in nature and can benefit from single-cycle read/write operations. Examples include chip status,
reset, sync options, checksum ramp parameters, and the page register.
GC5318 General Control
25
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
3.1 Control Data, Address, and Strobes
The control bus consists of 16 bidirectional control data lines C[0:15], six address lines A[0:5], a read
enable line RD, a write enable line WR, and a chip enable line CE. These lines usually interface to a
microprocessor or DSP chip and are intended to look like a block of memory.
Data is written by:
1. Setting up the desired address A[0:5]
2. Setting CE low,
3. Setting the desired data on C[0:15], and then
4. Pulsing WR low. Data is written when WR returns high.
3.2 MPU Timing Diagrams
3-Pin Mode (RD_N is the strobe)
WR_N
tUPCKL
tUPCKH
CE_N
RD_N
td(UP)
th
A(5:0)
D(15:0)
XXXX
VALID
2-Pin Mode (CE_N is the strobe, WR_N select direction)
RD_N is tied to GND
WR_N
tsu(UPA)
tUPCKL
tUPCKH
CE_N
td(UP)
th
A(5:0)
D(15:0)
XXXX
Figure 3-1. Read Diagrams
26
GC5318 General Control
VALID
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
3-Pin Mode (WR_N is the strobe)
RD_N
tUPCKL
tUPCKH
CE_N
WR_N
th(UPD)
A(5:0)
tsu(UPD)
D(15:0)
2-Pin Mode (CE_N is the strobe, WR_N select direction)
RD_N is tied to GND
WR_N
t3
tUPCKL
tUPCKH
CE_N
th(UPD)
A(5:0)
tsu(UPD)
D(15:0)
Figure 3-2. Write Diagrams
3.3 Interrupt Handling
When a GC5318 block sets an interrupt, the interrupt pin goes active if the interrupt source is not masked.
The microprocessor should then read the three interrupt flag registers to determine the source of the
interrupt. The microprocessor then has to read the interrupt source circuits register(s) to clear the interrupt
pin and interrupt flag register bit. The three interrupt registers are listed in the global registers part of the
control registers section.
3.4 Sync Signals
Various function blocks within the GC5318 need to be synchronized in order to realize predictable results.
The GC5318 provides a flexible system where each function block that requires synchronization can be
independently synchronized either from device pins or from a software one-shot. The one-shot option is
setup and triggered through control registers. There are four hardware sync input pins available. These
sync pins are qualified on the chip rising clock edge.
Table 3-1 shows the different sync modes available.
Table 3-1. Sync Modes Available
MODE
Transmit Sync Source
0
TxSyncA
1
TxSyncB
2
TxSyncC
3
TxSyncD
4
DUC sync counter TC
5
DUC sync triggered by one-shot
6
0 (always off)
7
1 (always on)
GC5318 General Control
27
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Table 3-2 and Table 3-3 summarize the blocks that have functions which can be synchronized using the
above eight sync options:
Table 3-2. Common Syncs
VARIABLE
ssel_comp_pmeter
ssel_duc_counter
DESCRIPTION
Initializes the xmit composite power meter
Initializes the xmit common sync counter
ssel_duc_serp
Initializes the xmit serial interface
ssel_duc_gain
Updates the gain register
ssel_duc_pilot
Initializes the xmit pilot generator and updates the pilot gain
ssel_duc_tadj
Updates the delay adjust register
ssel_duc_pmeter
Initializes the xmit channel power meter
Table 3-3. Channel Syncs
VARIABLE
DESCRIPTION
ssel_duc_nco
Resets the NCO accumulator
ssel_duc_freq
Updates the NCO frequency registers
ssel_duc_phase
Updates the NCO phase register
ssel_duc_dither
Resets the NCO dither
3.5 Initialization
Chip initialization procedures are available from Texas Instruments.
3.6 GC5318 Board Diagnostics
The GC5318 contains built-in test features that can be used to confirm that the chip is operating correctly
and to help users debug their boards and systems that contain the GC5318.
The diagnostic and board test procedures can be downloaded from the web at www.ti.com as the GC5318
Diagnostics Designer's Kit.
4 GC5318 Programming
The GC5318 contains over 3,000 control and coefficient registers, many of which must be initialized in
order to fully configure the chip. Rather than program each of these registers individually, Texas
Instruments supplies a configuration program called cmd5318 that accepts top-level configuration
information for the chip and then generates the full register map and control writes required to program the
chip.
The cmd5318 program, its user's guide, and example configuration files can be downloaded from the web
at www.ti.com as the GC5318 Configuration Designer's Kit.
The configuration controls have been defined in the functional description of each section of the chip. The
following tables summarize these controls and identify which register and which bits within the registers
they occupy.
Each control register table has a column identifying whether the variable must be specified by the user in
cmd5318 (U), is typically left at the default value and does not need to be specified (D), is computed by
cmd5318 and should not be set (C), or is for expert use only (E).
Tables are also included that show the top level page mapping for the chip controls, the status and
measurement result registers, and additional controls that are used with the GC101/GC5318 DIMM
evaluation platform.
28
GC5318 Programming
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
4.1 cmd5318 Keywords
These keywords are used by the cmd5318 program to set general configuration parameters.
Name
Argument
Use
Description
print
config
Global
Tells cmd5318 to generate a configuration output file for general use.
print
gc101
Global
Tells cmd5318 to generate a configuration output file for GC101 use.
print
analysis
Global
Tells cmd5318 to generate a analysis output file
print
table
Global
Tells cmd5318 to generate a table output file
print
power
Global
Tells cmd5318 to generate an approximate power consumption output
file.
txclk
clock frequency in MHz
Global
Used to calculate transmit tuning frequencies
duc
channel_number
DUC Channels
All controls after this keyword apply to this DUC channel
copy_ducchan
channel_number
DUC Channels
Copy the DUC channel commands from the specified channel to the
current channel
freqa
tuning frequency in MHz
DUCs
Sets the NCO tuning frequency for the UMTS channel or for the a-path
in the current channel if in CDMA mode.
freqb
tuning frequency in MHz
DUCs
Sets the NCO tuning frequency for the b-path in the current channel if in
CDMA mode.
pfir_coeff
filename for pfir taps
DUCs
Specifies the filename containing the pfir taps
cfir_coeff
filename for cfir taps
DUCs
Specifies the filename containing the cfir taps
overall_gaina
overall channel gain
DUCs
Optional—Specifies the overall gain for the UMTS channel or the a-path
in the current channel if in CDMA mode.
overall_gainb
overall channel gain
DUCs
Optional—Specifies the overall gain for the b-path in the current channel
if in CDMA mode.
4.1.1
GC5318 DIMM Keywords
These keywords are used to control how the GC5318 DIMM operates in the GC101 evaluation board.
Name
Argument
Type
Default
Use
Description
loopback
0 or 1
G
0
GC5318 DIMM
EVM signal select control must be set to 0.
spin0
0 or 1
G
0
GC5318 DIMM
EVM signal select control-txin_[0:5]_[a:b] ports.
spin1
0 or 1
G
0
GC5318 DIMM
EVM signal select control-txin_[6:11]_[a:b] ports.
sigout0
0-3
G
3
GC5318 DIMM
00 = txout_a enabled, 01 – txout_c enabled, 10—not used, 11—not used.
sigout1
0-3
G
3
GC5318 DIMM
00 = txout_b enabled, 01 – txout_d enabled, 10—not used, 11—not used.
txout_lsb
0 or 1
G
0
GC5318 DIMM
When asserted, the 2 LSBs each of active txout are output to GC101, else
the various strobes/syncouts are output.
sel_syncout
0-3
G
3
GC5318 DIMM
(if txout_lsb = 0) selects which signals are output. 00-tx_sync_out +
tx_i_flag, 01–sync_tst + aflag_tst, 10-tx_sync_out0 + interrupt.
res_op_en
0-3
G
3
GC5318 DIMM
(if txout_lsb = 0) when 00, test 9 + test11 is output , test7 + test8 data is
output.
sel_clkout
0-3
G
0
GC5318 DIMM
Selects the output clock source. 00 – txclk_out, 01-clk_tst, 10-clkout+.
adcclk_set
0 or 1
G
0
GC5318 DIMM
Not used.
GC5318 Programming
29
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
4.1.2
Page Map
This page map describes which pages and what registers within the pages are used. All other pages are
unused. This table is provided for reference only. The registers and the bits within the registers are
described in the following control register tables.
Pages
Address
Description
BASE+000 and BASE+020
00–1F
PFIR Coefficients, 2 LSBs
BASE+040 and BASE+060
00–1F
PFIR Coefficients, 16 MSBs
BASE+080 and BASE+0A0
00–1F
CFIR Coefficients, 2 LSBs
BASE+040 and BASE+0E0
00–1F
CFIR Coefficients, 16 MSBs
BASE+100
00–1F
Channel Control Registers
BASE+120
00–1D
Channel Control Registers
Where BASE is (DUCn x 0200) for DUCn from 0 to 11, and is (DDCn x 0200 + 2000) for DDCn from 0 to 11
1800
00–09
Reserved
1800
0A–1F
Reserved
1820
00–04
Reserved
1820
06
Reserved
1C00
00–11 and 1A–1E
General Transmit Control Registers
1C20
00–07 and 0C
General Transmit Control Registers
4.1.3
Status and Read-Only Registers
These registers can be accessed by the user to read status or read measurement results from the chip.
These register names are not used in cmd5318.
Name
Page
Address
LSB
Position
Bit
Width
Version
Global
20
0
5
A 5-bit read only register indicating the current GC5318 revision
status
inter_pmeter
Global
25
12
4
Indicates which transmit composite power meter generated the
interrupt
inter_tx_pmeter
Global
26
4
12
Indicates which transmit power meter generated the interrupt
inter_tx_cic
Global
28
0
12
Indicates which transmit cic overflow detect generated the interrupt
Description
comp_pmeter0_lsb
1C20
00
0
16
16 LSBs of composite power meter 0
comp_pmeter0_msb
1C20
01
0
16
16 MSBs of composite power meter 0
comp_pmeter1_lsb
1C20
02
0
16
16 LSBs of composite power meter 1
comp_pmeter1_msb
1C20
03
0
16
16 MSBs of composite power meter 1
comp_pmeter2_lsb
1C20
04
0
16
16 LSBs of composite power meter 2
comp_pmeter2_msb
1C20
05
0
16
16 MSBs of composite power meter 2
comp_pmeter3_lsb
1C20
06
0
16
16 LSBs of composite power meter 3
comp_pmeter3_msb
1C20
07
0
16
16 MSBs of composite power meter 3
tx_chk_sum
1C20
0C
0
16
Transmit checksum result
30
GC5318 Programming
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
4.1.4
Global Control Variables
These registers contain global controls for the GC5318. These registers are not paged and are accessed
directly using addresses 32–63 (20–3f hex)
Variable Name
Type
Address
LSB
Position
Bit
Width
Default
page
E
21
0
16
0
The page register selects which page addresses 00-1F
(0-31) will access.
slf_tst_ena
D
22
15
1
0
(TESTING PURPOSES) Turns on the checksum LFSR.
tst_sel_chan
E
22
1
2
0
(TESTING PURPOSES) In each slice, these bits control
which tst_out is sent to the transmit block. (which duc in the
slice)
tst_on
E
22
0
1
0
(TESTING PURPOSES) When asserted the testbus is
active; txout_c (17:0), and txout_d (17:0) form the 36-bit test
word output.
Description
The following tristates are active low; 0 turns the output on; 1
tristates it.
3-state_10
E
23
10
1
1
Reserved outputs for test; must be set to 1 (tristate)
3-state_9
C
23
9
1
1
This bit turns on the slice5 tx_sync.
3-state_8
C
23
8
1
1
This bit turns on the slice4 tx_sync.
3-state_7
C
23
7
1
1
This bit turns on the slice3 tx_sync.
3-state_6
C
23
6
1
1
This bit turns on the slice2 tx_sync.
3-state_5
C
23
5
1
1
This bit turns on the slice1 tx_sync.
3-state_4
C
23
4
1
1
This bit turns on the slice0 tx_syncand tx_sync_out.
3-state_3
C
23
3
1
1
This turns on the txout_d outputs.
3-state_2
C
23
2
1
1
This turns on the txout_c CLK_TST, IFLAG_TST, and
SYNC_TST outputs.
3-state_1
C
23
1
1
1
This turns on the txout_b outputs.
3-state_0
C
23
0
1
1
This turns on the txout_a, TX_IFLAG, and TXCLK_OUT
outputs.
tx_oneshot
D
24
15
1
0
When set, a one-shot pulse is sent to the transmit blocks for
syncing. This option works only if the blocks are programmed
to see the oneshot. To use the one-shot again, it must be
programmed back to a '0' and then back to a '1'.
imask_comp_pmeter
D
29
12
4
0
Interrupt mask bits for the transmit composite power meter
imask_tx_pmeter
D
2A
4
12
0
Interrupt mask bits for the composite power meter
imask_tx_cic
D
2C
0
12
0
Interrupt mask bits for overflow detection in the transmit cics
GC5318 Programming
31
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
4.1.5
General Controls
These registers control the transmit output interface from the channels.
Variable Name
Type
Page
Address
LSB
Position
Bit
Width
Default
tst_sel_slice
E
1C00
00
13
3
0
(TESTING PURPOSES) This value selects the
slice block that is generating the tst_out data.
(Which DUC)
tst_rate
E
1C00
05
11
5
0
The value here controls the output clock rate on
the clk_tst pin. A value of 0 gives a full-rate
output clock (txclk rate), a 1 gives half-rate
output clock, a 3 gives 1/4th rate output clock,
and so on. The number of txclk cycles for which
the clk_tst signal is high + low = 1 + tst_rate.
interf_round
D
1C00
00
8
2
0
Controls round point on the transmit output data;
{00 = 18b, 01 = 16b, 10 = 14b, 11 = 12b}.
Rounded output data is MSB justified. For
example, a 12b round point causes the output
data to be presented on the output pins (17:6),
and the output pins (5:0) to be held low.
interf_ena
D
1C00
00
4
4
15
Enables the individual transmit output busses 3
through 0. Disabled busses are always held low.
interf_interl
U
1C00
00
1
1
0
Enables interleaved I/Q data when asserted.
interf_real
U
1C00
00
0
1
1
Enables real only outputs when asserted.
Complex data is output when cleared.
interf_scale_3
U
1C00
01
12
4
0
Selects the scaling between the sumchain
output signals and the transmit output pins and
transmit composite power meters. Appropriate
limiting and rounding is performed as required
by the programmed round point. Gain =
2(interf_scale). For sumchain 3.
interf_scale_2
U
1C00
01
8
4
0
Selects the scaling between the sumchain
output signals and the transmit output pins and
transmit composite power meters. Appropriate
limiting and rounding is performed as required
by the programmed round point. Gain =
2(interf_scale). For sumchain 2
interf_scale_1
U
1C00
01
4
4
0
Selects the scaling between the sumchain
output signals and the transmit output pins and
transmit composite power meters. Appropriate
limiting and rounding is performed as required
by the programmed round point. Gain =
2(interf_scale). For sumchain 1
interf_scale_0
U
1C00
01
0
4
0
Selects the scaling between the sumchain
output signals and the transmit output pins and
transmit composite power meters. Appropriate
limiting and rounding is performed as required
by the programmed round point. Gain =
2(interf_scale). For sumchain 0
comp_pmeter0_count_lsb
D
1C00
02
0
16
0
This is the number of sample sets to accumulate
for a power measurement. Ia and Qa (signal)
are each squared and accumulated. Each pair
of I and Q are equal to one integration count.
The accumulation interval is initiated when the
sync is asserted and the programmed
sync_delay has expired or when the interval
start time is reached. When the integration count
is reached, the accumulated powers are made
available for MPU access and an interrupt is
generated. Bits 0–15
comp_pmeter0_count_msb
D
1C00
03
0
5
0
Bits 16–20 of the number of sample sets to
accumulate for a power measurement. (Used in
conjunction with the previous variable.)
comp_pmeter0_sync_delay
D
1C00
03
7
9
0
Programmable start delay from sync, in eight
output sample units.
32
GC5318 Programming
Description
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Variable Name
Type
Page
Address
LSB
Position
Bit
Width
Default
comp_pmeter0_interval_lsb
D
1C00
04
0
16
0
This is the interval over which the integration is
restarted and must be greater than the
integration count. The interval start counter and
RMS power accumulation is started at the sync
pulse after the programmed delay and every
time the interval counter reaches its limit. Bits
0–15
comp_pmeter0_interval_msb
D
1C00
05
0
5
0
Bits 16–20 of the interval over which the
integration is restarted. (Used in conjunction with
the previous variable.)
Description
comp_pmeter1_count_lsb
D
1C00
06
0
16
0
See description for pmeter0
comp_pmeter1_count_msb
D
1C00
07
0
5
0
See description for pmeter0
comp_pmeter1_sync_delay
D
1C00
07
7
9
0
See description for pmeter0
comp_pmeter1_interval_lsb
D
1C00
08
0
16
0
See description for pmeter0
comp_pmeter1_interval_msb
D
1C00
09
0
5
0
See description for pmeter0
comp_pmeter2_count_lsb
D
1C00
0A
0
16
0
See description for pmeter0
comp_pmeter2_count_msb
D
1C00
0B
0
5
0
See description for pmeter0
comp_pmeter2_sync_delay
D
1C00
0B
7
9
0
See description for pmeter0
comp_pmeter2_interval_lsb
D
1C00
0C
0
16
0
See description for pmeter0
comp_pmeter2_interval_msb
D
1C00
0D
0
5
0
See description for pmeter0
comp_pmeter3_count_lsb
D
1C00
0E
0
16
0
See description for pmeter0
comp_pmeter3_count_msb
D
1C00
0F
0
5
0
See description for pmeter0
comp_pmeter3_sync_delay
D
1C00
0F
7
9
0
See description for pmeter0
comp_pmeter3_interval_lsb
D
1C00
10
0
16
0
See description for pmeter0
comp_pmeter3_interval_msb
D
1C00
11
0
5
0
See description for pmeter0
duc_counter_lsb
D
1C00
1A
0
16
65535
32-bit interval timer common to all DUC sync
inputs. This timer may be programmed to any
interval count, and each DUC synchronization
input can select this counter as a source. This
counter increments on every TXCLK rising edge.
Bits 0–15
duc_counter_msb
D
1C00
1B
0
16
65535
Bits 16-31 of the above mentioned 32-bit interval
timer.
ssel_duc_counter
U
1C00
1C
8
3
0
Selects the sync source for the DUC sync
counter.
duc_counter_width
D
1C00
1C
0
8
0
Sets the width of the counter generated sync
pulse in TX clock cycles, from 1 to 256. The
width of this pulse must be long enough to be
captured by the slowest block to use the DUC
counter sync.
ssel_comp_pmeter_0
U
1C00
1D
12
3
0
Selects the sync source for composite power
meter 0.
ssel_comp_pmeter_1
U
1C00
1D
8
3
0
Selects the sync source for composite power
meter 1.
ssel_comp_pmeter_2
U
1C00
1D
4
3
0
Selects the sync source for composite power
meter 2.
ssel_comp_pmeter_3
U
1C00
1D
0
3
0
Selects the sync source for composite power
meter 3.
ssel_txsync_out
U
1C00
1E
0
3
0
Selects the sync source for the TXSYNC_OUT
pin.
GC5318 Programming
33
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
4.1.6
DUC Channel Controls
These controls follow the duc <channel_number> keywords in the cmd5318 configuration file.
Variable Name
Address
(HEX)
LSB
Position
Bit
Width
Default
Description
cdma_mode
U
0100
00
15
1
1
When asserted the block is in the dual channel CDMA2000
mode.
crastarttap_pfir
C
0100
00
8
5
0
These bits define the number of taps that PFIR uses for the
filtering. Another way of looking at these bits is that this
value is the location in the RAM of the center tap. PFIR: (2
x crastarttap_pfir) +1. Note: crastarttap_pfir must be odd.
crastarttap_cfir
C
0100
00
3
5
0
These bits define the number of taps that CFIR uses for the
filtering. CFIR: (2 x crastarttap_cfir) +1. Note:
crastarttap_cfir must be odd.
symmetric_pfir
C
0100
00
14
1
0
When asserted the block PFIR is symmetric.
symmetric_cfir
C
0100
00
13
1
0
When asserted the block CFIR is symmetric.
pfir_gain
U
0100
01
13
3
0
this is the gain for the PFIR. 0 = 2e – 18, 1= 2e – 17.
cfir_gain
U
0100
01
5
1
0
This is the gain for the CFIR. 0 = 2e – 19, 1 = 2e – 18.
pmeter_integration_du
c
D
0100
02
0
13
0
This is the number of four sample sets to accumulate for a
power measurement. In CDMA mode, one sample set is
the I and Q of the signal and diversity. Ia and Qa (signal)
are each squared and accumulated and Ib and Qb
(diversity) are squared and accumulated. In UMTS mode,
each I and Q pair are squared and accumulated. Four
samples are equal to one integration count. The count is
initiated when the sync is asserted or when the interval
start time is reached. When the integration count is
reached, the accumulated powers are made available for
MPU access and an interrupt is generated.
pmeter_sync_delay_d
uc
D
0100
03
9
7
0
The delay from selected sync source to when the power
calculation starts.
pmeter_interval_duc
D
0100
03
0
9
0
The start interval timer is the interval over which the
integration is restarted and must be greater than the
integration count. The interval start counter and RMS
power accumulation is started at the sync pulse after the
programmed delay and every time the interval counter
reaches its limit. This value is in 64 sample units.
pilot_gain_0
D
0100
04
0
16
0
Pilot channel gain word, aligned with MSB of the input data.
0xFFFF generates a full scale complex pilot signal added to
the user signal. Setting the gain to 0x0000 causes no pilot
signal to be added. Only valid for UMTS, should be set to
0x0000 for CDMA.
34
Type Page
pilot_gain_1
D
0100
05
0
16
0
This value MUST be set to the same value as pilot_gain_0.
pilot_psc_lsb
D
0100
06
0
16
1
The lower 16 bits of the 18-bit pilot X LFSR initial value.
This 18b word is loaded on pilot sync event. The value
loaded here that corresponds to 3gpp primary scrambling
code (PSC) 0 is 0x00001. Users must calculate the correct
initial value to implement the other 511 PSCs.
pilot_psc_msb
D
0100
07
14
2
0
The upper 2 bits of the 18-bit pilot X LFSR initial value. This
18b word is loaded on pilot sync events. The value loaded
here that corresponds to 3gpp primary scrambling code
(PSC) 0 is 0x00001. Users must calculate the correct initial
value to implement the other 511 PSCs.
pilot_diversity
D
0100
07
13
1
0
Select between main and diversity pilot symbol generation.
(0 = main, 1 = diversity)
pilot_delay
D
0100
08
0
16
0
Unsigned delay value in chips from the pilot sync event,
from 0 to 38399 chips.
ssel_pmeter
U
0120
0B
8
3
0
Selects the sync source for the channel power meter.
ssel_serial
U
0120
0B
0
3
0
Selects the sync source for the DUC serial interface state
machines.
ssel_tadj_fine
U
0120
0C
12
3
0
Selects the sync source for the fine time adjust decimation
(DUC).
GC5318 Programming
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Variable Name
Type Page
Address
(HEX)
LSB
Position
Bit
Width
Default
3
0
Selects the sync source for the course time adjust delay
selection.
Selects the sync source for the DUC gain register.
Description
ssel_tadj_coarse
U
0120
0C
8
ssel_gain
U
0120
0C
4
3
0
gainfora
U
0100
0C
0
16
8192
This is the unsigned gain that is multiplied with the CDMA
channel A or UMTS channel input signal. The gain multiply
is calculated as gainfora/8192.
gainforb
U
0100
0D
0
16
8192
This is the unsigned gain that is multiplied with the CDMA
channel B input signal.
ssel_nco
U
0120
0D
12
3
0
Selects the sync source for the NCO accumulator reset.
ssel_dither
U
0120
0D
8
3
0
Selects the sync source for the NCO phase dither
generator reset.
ssel_freq
U
0120
0D
4
3
0
Selects the sync source for the NCO frequency register.
ssel_phase
U
0120
0D
0
3
0
Selects the sync source for the NCO phase offset register.
cic_scale_a
U
0100
0E
11
5
0
This sets the gain shift at the output of the CDMA A
channel (or UMTS channel) CIC. 0x00 is no shift; each
increment by 1 increases the signal amplitude by 2X.
cic_scale_b
U
0100
0E
6
5
0
This sets the gain shift at the output of the CDMA B
channel CIC. 0x00 is no shift; each increment by 1
increases the signal amplitude by 2X.
cic_interp_decim
U
0100
0E
0
5
24
Sets the CIC interpolation, where interpolation is
cic_interp_decim + 1 in the digital up converters.
cic_m2_ena_a
D
0100
0F
10
6
0
Programs the CDMA A channel (or UMTS channel) CIC fir
sections M value to 2 when set, 1 when cleared.
cic_m2_ena_a(0) controls the M value for the first comb
section and cic_m2_ena_a(5) controls the M value for the
last comb section.
cic_m2_ena_b
D
0100
0F
4
6
0
Programs the CDMA B channel CIC fir sections M value to
2 when set, 1 when cleared. cic_m2_ena_b(0) controls the
M value for the first comb section and cic_m2_ena_b(5)
controls the M value for the last comb section.
cic_auto_flush_dis
E
0100
10
12
4
0
Disables the automatic flush feature in the CIC
accumulators.
cic_flush_test
E
0100
10
8
4
0
Forces an overflow detection in the CIC only on a rising
edge of this bit, therefore it must be programmed to '0' and
then back to `1' for the edge to occur.
cic_flush_clear
E
0100
10
4
4
0
Clears an overflow error manually when set, again only on
a rising edge does this occur.
tadj_offset_coarse_a
D
0100
11
10
6
0
This is part of the time delay adjust. This is the coarse
offset and is really an offset from the write address in the
delay ram. This value affects the A channel if CDMA mode
is being used, or the UMTS channel. Each LSB is one
more offset between input to the course delay block and
the output of the course block.
tadj_offset_coarse_b
D
0100
11
4
6
0
This is part of the time delay adjust. This is the coarse
offset and is really an offset from the write address in the
delay ram. This value affects the B channel if CDMA mode
is being used. Each LSB is one more offset between input
to the course delay block and the output of the course
block.
tadj_offset_fine_a
D
0100
12
13
3
0
This is part of the time delay adjust. This is the fine adjust
value. It adjusts the time delay at the clock rate. This value
affects the A channel if CDMA mode is being used, or the
UMTS channel.
tadj_offset_fine_b
D
0100
12
10
3
0
This is part of the time delay adjust. This is the fine adjust
value. It adjusts the time delay at the clock rate. This value
affects the B channel if CDMA mode is being used.
tadj_interp_decim
U
0100
12
7
3
1
This is the decimation or interpolation value for the fine
time adjust block. Decimation or interpolation can be from 1
to 8. This value affects both the A and B channels if CDMA
mode is being used, or the UMTS channel.
GC5318 Programming
35
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Variable Name
Type Page
Address
(HEX)
LSB
Position
Bit
Width
Default
Description
phase_add_a_lsb
C
0100
13
0
16
0
This 32 bit word is used to control the frequency of the
NCO. Derived from the keyword freqa by cmd5318. (for
CDMA channel A or UMTS channel). Lower 16 bits.
phase_add_a_msb
C
0100
14
0
16
0
Upper 16 bits of the above 32-bit word.
phase_add_b_lsb
C
0100
15
0
16
0
This 32-bit word is used to control the frequency of the
NCO. Derived from the keyword freqb by cmd5318. (for
CDMA channel B). Lower 16 bits.
phase_add_b_msb
C
0100
16
0
16
0
Upper 16 bits of the above 32-bit word.
phase_offset_a
D
0100
17
0
16
0
This is the fixed phase offset added to the output of the
frequency accumulator for sinusoid generation in the NCO.
(UMTS mode and A channel in CDMA mode)
phase_offset_b
D
0100
18
0
16
0
This is the fixed phase offset added to the output of the
frequency accumulator for sinusoid generation in the NCO
for CDMA B channel.
dither_ena
D
0100
19
15
1
0
This bit controls whether or not dither is turned on(1) or
off(0).
test_bits_1
E
0100
19
13
2
0
TEST BITS. Set to 0 for normal operation.
pmeter_sync_disable
D
0100
19
12
1
0
Turns off the sync to the channel power meter. This can be
used to individually turn off syncs to a channels power
meter, while still having syncs to other power meters on the
chip.
ddc_duc_ena
U
0100
19
11
1
0
When set this turns on the DUC.
mixer_gain
U
0100
19
9
1
0
Adds a fixed –6 dB of gain to the mixer output(before round
and limiting) when asserted. Else adds –12-dB gain when
deasserted.
mpu_ram_read
E
0100
19
8
1
0
(TESTING PURPOSES) Allows the coefficient RAMs in the
PFIR/CFIR to be read out the mpu data bus. This cannot
be done during normal operation and must be done when
the state of the output data is not important. THIS BIT
MUST BE SET ONLY DURING THE READ OPERATION.
sumchn_sel_b
U
0100
19
4
4
2
This word controls the second set of additions for the
CDMA B signal in the sumchn output. The selection bits
are not mutually exclusive.
sumchn_sel_a
U
0100
19
0
4
1
This word controls the first set of additions for the CDMA A
signal (or UMTS signal) in the sumchn output. The
selection bits are not mutually exclusive.
tst_sel_block
E
0100
1A
0
6
0
(TESTING PURPOSES) This is the selection of which
signal comes out the test bus. When a constant `0' is
selected this also reduces power by preventing the data at
the input of the test block from changing. Howeve,rit does
not stop the clock.
serp_tran_bits
U
0100
1B
11
5
17
Number of input bits per sample-1; for 18 bits, this is set to
{10001}.
serp_tran_fsdel
D
0100
1B
8
2
1
Delay between frame sync output and MSB of serial data
{3, 2, 1, 0}.
serp_tran_4pin
D
0100
1B
7
1
0
Selects 2-pin mode when cleared and 4-pin mode when
set.
36
serp_tran_fsinvl
U
0100
1B
0
7
50
Transmit serial interface frame sync interval in bit clocks.
serp_tran_clkdiv
U
0100
1C
0
4
1
Transmit serial interface clock divider rate-1; 0 is full rate,
and 15 divides the clock by 16. For example, to run the
serial interface at 1/4 the transmit clock, set
serp_tran_clkdiv(3:0) = 0011.
ssel_pilot
U
0120
0B
4
3
0
Selects the sync source for the DUC pilot code generator.
GC5318 Programming
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
5 GC5318 Pin Description
5.1 Signals
Signal Name
Ball Desig
Type
txclk
K26
input
Transmit clock input
Description
txin_0_a
T23
input
DUC 0 serial in data. CDMA A: I/Q UMTS: I
txin_1_a
U25
input
DUC 1 serial in data. CDMA A: I/Q UMTS: I
txin_0_b
T24
input
DUC 0 serial in data. CDMA B: I/Q UMTS: Q
txin_1_b
U26
input
DUC 1 serial in data. CDMA B: I/Q UMTS: Q
txin_2_a
W26
input
DUC 2 serial in data. CDMA A: I/Q UMTS: I
txin_3_a
V25
input
DUC 3 serial in data. CDMA A: I/Q UMTS: I
txin_2_b
U24
input
DUC 2 serial in data. CDMA B: I/Q UMTS: Q
txin_3_b
V26
input
DUC 3 serial in data. CDMA B: I/Q UMTS: Q
txin_4_a
Y26
input
DUC 4 serial in data. CDMA A: I/Q UMTS: I
txin_5_a
W25
input
DUC 5 serial in data. CDMA A: I/Q UMTS: I
txin_4_b
V24
input
DUC 4 serial in data. CDMA B: I/Q UMTS: Q
txin_5_b
U23
input
DUC 5 serial in data. CDMA B: I/Q UMTS: Q
txin_6_a
W23
input
DUC 6 serial in data. CDMA A: I/Q UMTS: I
txin_7_a
AA26
input
DUC 7 serial in data. CDMA A: I/Q UMTS: I
txin_6_b
Y25
input
DUC 6 serial in data. CDMA B: I/Q UMTS: Q
txin_7_b
W24
input
DUC 7 serial in data. CDMA B: I/Q UMTS: Q
txin_8_a
Y23
input
DUC 8 serial in data. CDMA A: I/Q UMTS: I
txin_9_a
AB26
input
DUC 9 serial in data. CDMA A: I/Q UMTS: I
txin_8_b
AA25
input
DUC 8 serial in data. CDMA B: I/Q UMTS: Q
txin_9_b
Y24
input
DUC 9 serial in data. CDMA B: I/Q UMTS: Q
txin_10_a
AA23
input
DUC 10 serial in data. CDMA A: I/Q UMTS: I
txin_11_a
AC26
input
DUC 11 serial in data. CDMA A: I/Q UMTS: I
txin_10_b
AB25
input
DUC 10 serial in data. CDMA B: I/Q UMTS: Q
txin_11_b
AA24
input
DUC 11 serial in data. CDMA B: I/Q UMTS: Q
tx_sync_out_0
AB24
output
Transmit serial interface strobe for DUC 0, 1 (txin_[0,1]_[a,b])
tx_sync_out_1
AC25
output
Transmit serial interface strobe for DUC 2, 3 (txin_[2,3]_[a,b])
tx_sync_out_2
AD26
output
Transmit serial interface strobe for DUC 4, 5 (txin_[4,5]_[a,b])
tx_sync_out_3
AB23
output
Transmit serial interface strobe for DUC 6, 7 (txin_[6,7]_[a,b])
tx_sync_out_4
AC24
output
Transmit serial interface strobe for DUC 8, 9 (txin_[8,9]_[a,b])
tx_sync_out_5
AD23
output
Transmit serial interface strobe for DUC 10, 11 (txin_[10,11]_[a,b])
tx_synca
K24
input
Transmit sync input
tx_syncb
J25
input
Transmit sync input
tx_syncc
H26
input
Transmit sync input
tx_syncd
K23
input
Transmit sync input
tx_sync_out
F23
output
Transmit general purpose output sync
txclk_out
E23
output
Transmit output clock
tx_i_flag
D24
output
Transmit output iflag
txout_a_17
B19
output
Transmit output bus a MSB
txout_a_16
A20
output
Transmit output bus a
txout_a_15
C19
output
Transmit output bus a
txout_a_14
B20
output
Transmit output bus a
txout_a_13
A21
output
Transmit output bus a
txout_a_12
D19
output
Transmit output bus a
txout_a_11
C20
output
Transmit output bus a
GC5318 Pin Description
37
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
38
Signal Name
Ball Desig
Type
txout_a_10
B21
output
Transmit output bus a
txout_a_9
A22
output
Transmit output bus a
txout_a_8
D20
output
Transmit output bus a
txout_a_7
C21
output
Transmit output bus a
txout_a_6
B22
output
Transmit output bus a
txout_a_5
A23
output
Transmit output bus a
txout_a_4
C22
output
Transmit output bus a
txout_a_3
B23
output
Transmit output bus a
txout_a_2
A24
output
Transmit output bus a
txout_a_1
D22
output
Transmit output bus a
txout_a_0
C23
output
Transmit output bus a LSB
txout_b_17
B14
output
Transmit output bus b MSB
txout_b_16
C14
output
Transmit output bus b
txout_b_15
D14
output
Transmit output bus b
txout_b_14
A15
output
Transmit output bus b
txout_b_13
B15
output
Transmit output bus b
txout_b_12
C15
output
Transmit output bus b
txout_b_11
A16
output
Transmit output bus b
txout_b_10
B16
output
Transmit output bus b
txout_b_9
A17
output
Transmit output bus b
txout_b_8
C16
output
Transmit output bus b
txout_b_7
B17
output
Transmit output bus b
txout_b_6
D16
output
Transmit output bus b
txout_b_5
A18
output
Transmit output bus b
txout_b_4
C17
output
Transmit output bus b
txout_b_3
B18
output
Transmit output bus b
txout_b_2
A19
output
Transmit output bus b
txout_b_1
D17
output
Transmit output bus b
txout_b_0
C18
output
Transmit output bus b LSB
txout_c_17
C9
output
Transmit output bus c MSB
txout_c_16
D10
output
Transmit output bus c
txout_c_15
A8
output
Transmit output bus c
txout_c_14
B9
output
Transmit output bus c
txout_c_13
C10
output
Transmit output bus c
txout_c_12
A9
output
Transmit output bus c
txout_c_11
D11
output
Transmit output bus c
txout_c_10
B10
output
Transmit output bus c
txout_c_9
C11
output
Transmit output bus c
txout_c_8
A10
output
Transmit output bus c
txout_c_7
B11
output
Transmit output bus c
txout_c_6
A11
output
Transmit output bus c
txout_c_5
C12
output
Transmit output bus c
txout_c_4
B12
output
Transmit output bus c
txout_c_3
A12
output
Transmit output bus c
txout_c_2
D13
output
Transmit output bus c
txout_c_1
C13
output
Transmit output bus c
txout_c_0
B13
output
Transmit output bus c LSB
txout_d_17
C4
output
Transmit output bus d MSB
GC5318 Pin Description
Description
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
Signal Name
Ball Desig
Type
txout_d_16
D5
output
Transmit output bus d
Description
txout_d_15
A3
output
Transmit output bus d
txout_d_14
B4
output
Transmit output bus d
txout_d_13
C5
output
Transmit output bus d
txout_d_12
A4
output
Transmit output bus d
txout_d_11
B5
output
Transmit output bus d
txout_d_10
C6
output
Transmit output bus d
txout_d_9
D7
output
Transmit output bus d
txout_d_8
A5
output
Transmit output bus d
txout_d_7
B6
output
Transmit output bus d
txout_d_6
C7
output
Transmit output bus d
txout_d_5
D8
output
Transmit output bus d
txout_d_4
A6
output
Transmit output bus d
txout_d_3
B7
output
Transmit output bus d
txout_d_2
C8
output
Transmit output bus d
txout_d_1
A7
output
Transmit output bus d
txout_d_0
B8
output
Transmit output bus d LSB
5.2 Microprocessor Signals
Signal Name
Ball Desig
Type
d0
AD8
input/output
MPU register interface data bus LSB
Description
d1
AE7
input/output
MPU register interface data bus
d2
AF6
input/output
MPU register interface data bus
d3
AC8
input/output
MPU register interface data bus
d4
AD7
input/output
MPU register interface data bus
d5
AE6
input/output
MPU register interface data bus
d6
AF5
input/output
MPU register interface data bus
d7
AC7
input/output
MPU register interface data bus
d8
AD6
input/output
MPU register interface data bus
d9
AE5
input/output
MPU register interface data bus
d10
AF4
input/output
MPU register interface data bus
d11
AD5
input/output
MPU register interface data bus
d12
AE4
input/output
MPU register interface data bus
d13
AF3
input/output
MPU register interface data bus
d14
AC5
input/output
MPU register interface data bus
d15
AD4
input/output
MPU register interface data bus MSB
a0
AB4
input
MPU register interface address bus LSB
a1
AD1
input
MPU register interface address bus
a2
AC2
input
MPU register interface address bus
a3
AB3
input
MPU register interface address bus
a4
AA4
input
MPU register interface address bus
a5
AC1
input
MPU register interface address bus MSB
rd_n
Y4
input
MPU register interface read—active low
wr_n
AA3
input
MPU register interface write—active low
ce_n
AB2
input
MPU register interface chip enable—active low
reset_n
R24
input
Chip reset—active low
interrupt
AC3
output
Chip interrupt
GC5318 Pin Description
39
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
5.3 JTAG Signals
Signal Name
Ball Desig
Type
tdi
N23
input
JTAG test data in
Description
tms
M26
input
JTAG test mode select
trst_n
M25
input
JTAG test reset (same as trst—the "_n" is for consistency, being active low)
tck
M24
input
JTAG test clock
tdo
L26
output
JTAG test data out
5.4 Factory Test and No Connect Signals
Signal Name
Ball Desig
Type
testmode0
R26
Input
Do not connect
Description
testmode1
P24
Input
Do not connect
scanen
P25
Input
Do not connect
aflag_tst
E24
Output
Do not connect
sync_tst
D25
Output
Do not connect
clk_tst
C26
Output
Do not connect
fa002_scan
T26
Input
Do not connect
fa002_clk
R23
Input
Do not connect
fa002_out
T25
Output
Do not connect
zero
N25
Input
Do not connect
rxclk
L24
Input
Ground
adcclk
D3
Input
Ground
F26, G24, G25, H23, L23
Input
Tie each pin high through 100Ω resistors to VPAD
K25, M23, L25, N24, R25, D26, E25, E26, F24,
F25, G23, J26
No connect
Do not connect
C1, D1, D2, D3, E1–E4, F1–F4, G1– G4, G26,
H1–H4, H24, H25
No connect
Do not connect
J1–J3, K1, J24, K2–K4, L1– L4, L24, M1–M4,
N2, N3, P2–P4
No connect
Do not connect
R1– R4, T1–T4, U1– U4, V1–V3, W1– W4,
Y1–Y3
No connect
Do not connect
AA1, AA2, AB1, AC10, AC11, AC13, AC14,
AC16, AC17, AC19, AC20, AC22, AD9–AD22
No connect
Do not connect
AE8–AE23, AF7–AF12, AF15–AF24
No connect
Do not connect
5.5 Power and Ground Signals
Signal Name
GND
Ball Desig
Description
A1, A2, A13, A14, A25, A26, B1, B3, B24, B26, C2, C25, N1, N26, P1, P26, AD2, AD25, AE1, AE24, Ground
AE3, AE26, AF1, AF2, AF13, AF14, AF25, AF26, L11, L12, L13, L14, L15, L16, M11–M16, N11–N16,
P11–P16, R11–R16, T11–T16
VCORE
B2, D4, N4, AC4, AE2, B25, D23, P23, AC23, AE25, C3, J4, V4, AD3, C24, J23, V23, AD24
VPAD
D6, D12, D18, AC6, AC12, AC18, D9, D15, D21, AC9, AC15, AC21
Core power
I/O power
5.6 Power Monitoring
40
Signal Name
Ball Desig
vcoremon
N24
These pins monitor the internal power distribution. They cannot carry significant current and should not
be connected to normal power and ground. It is recommended that this pin be brought to a small probe
point for future monitoring/debugging purposes.
gndmon
R25
It is recommended that this pin be brought to a probe point for future monitoring/debugging purposes.
GC5318 Pin Description
Description
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
5.7 JTAG
The JTAG standard for boundary scan testing is implemented for board testing purposes. Internal scan
test is not be supported. Five device pins are dedicated for JTAG support: tdi, tdo, tms, tck, and trst_n.
The BSDL file is available on the web.
6 Specifications
6.1 Absolute Maximum Ratings (1)
VPAD
Pad ring supply voltage
VCORE
Core supply voltage
VIN
Input voltage (undershoot and overshoot)
VALUE
UNIT
–0.3 to +4
V
–0.3 to +1.8
V
–0.5 to VPAD + 0.5
V
Clamp current for an input or output
–20 to +20
mA
Tstg
Storage temperature
–65 to +140
°C
TJ
Junction temperature
+105
°C
+300
°C
Lead soldering temperature (10 seconds)
ESD classification
Class 2 (Passed 2.5-kV HBM, 500-V CDM, 150-V MM)
Moisture sensitivity
Class 4 (Four days' floor life at 30°C/60%H)
Reflow conditions
(1)
JEDEC standard, 240°C max
Exceeding the absolute maximum ratings (min or max) may cause permanent damage to the part. These are stress only ratings and are
not intended for operation.
6.2
Recommended Operating Conditions
MIN
VPAD
Pad ring supply voltage
VCORE
Core supply voltage
VPAD – VCORE
Supply voltage difference
TA
Temperature ambient, no air
TJ
Junction temperature (2)
(1)
(2)
flow (1)
MAX
UNIT
3
3.6
V
1.5
1.65
V
2
V
+85
°C
+105
°C
–40
Chips specifications in Section 6.3 and Section 6.5 are production tested to +100°C case temperature. QA tests are performed at
+85°C.
Thermal management will be required for full rate operation; see the following table. The circuit is designed for junction temperatures up
to +125°C. Sustained operation at elevated temperatures reduces long-term reliability. Lifetime calculations based on maximum junction
temperature of +105°C.
Specifications
41
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
6.3 Thermal Characteristics
THERMAL CONDUCTIVITY (1)
388 BGA
UNIT
θJA
Theta junction-to-ambient (still air)
13.5
°C/W
θJA2m
Theta junction-to-ambient (2m/s estimated)
9.3
°C/W
θJC
Theta junction-to-case
2.4
°C/W
(1)
Air flow reduces θJA and is highly recommended.
6.4 Power Consumption
The maximum power consumption is a function of the operating mode of the chip. The cmd5318 estimates
the typical power supply current for the chip in a specific configuration. The AC Characteristics table
provides maximum current in a maximum configuration used in production test.
Current consumption on the pad supply is primarily due to the external loads and follows C x V x F.
Internal loads are estimated at 2 pF per pin. Data outputs have a transition density of going from a zero to
a one, once per four clocks, while clock outputs transition every cycle. The frame strobes consume
negligible power due to the low transition frequency. In general:
Ipad = Σ DataPad/4 x C x F x V + Σ ClockPad x C x F x V
A worst case current would be all transmit ports operating at 125 MHz.
Ipad = (1+(4 x 18 + 4 x 2 x 6)/4) x (C + 2pF) x Fout x Vpad = 31 x 22 pF x 125 MHz x 3.3 V = 280 mA.
A more typical application with two ports active would use roughly 150 mA.
6.5 DC Operating Conditions
At –40°C to +85°C case, unless otherwise noted.
PARAMETER
VIL
Voltage input low (1) (2)
VIH
Voltage input high (1) (2)
VOL
Voltage output low (1) (2) (IOL = 2 mA)
high (1) (2)
VPAD = 3 V to 3.6 V
MIN
TYP
MAX
0.8
2
V
V
0.5
V
VOH
Voltage output
2.4
VPAD
V
| IPU |
Pullup current (VIN = 0 V) (tdi, tms, trst_n, reset_n) (nominal 20 µA) (1) (2)
5
35
µA
| IPD |
Pulldown current (VIN = VPAD) (all other inputs and bidirs) (nominal 20 µA) (2)
5
35
µA
| IIN |
ICCQ
(IOH = –2 mA)
UNIT
Leakage (VIN = VPAD) (tdi, tms, trst_n, reset_n) (2)
2
Leakage (VIN = 0) (all other inputs and bidirs) (2)
2
Leakage (VIN = 0 or VPAD) (all outputs) (2)
2
Quiescent supply current, ICORE (2)
8
inputs (3)
CIN
Capacitance for
CBI
Capacitance for bidirectionals (3)
(1)
(2)
(3)
42
Voltages are measured at low speed. Output voltages are measured with the indicated current load.
Each part is tested at +100°C case temperature for the given specification. Lots are sample tested at –40°C.
Controlled by design and process and not directly tested.
Specifications
µA
mA
5
pF
5
pF
GC5318
HIGH-DENSITY DIGITAL UPCONVERTER
www.ti.com
SLWS187 – JULY 2006
6.6 AC Characteristics
PARAMETER (1)
MIN
Clock frequency ( txclk) in selected modes (2) (3)
FCK
MAX
125
Clock frequency ( txclk) unrestricted (2)
80
UNIT
MHz
tTXCKL
Clock low period (below VIL) (txclk) (2)
3
ns
tTXCKH
Clock high period (above VIH) (txclk) (2)
3
ns
(txclk) (4)
tr, tf
Clock rise and fall times (VIL to VIH)
tsu(TX)
Input setup (txin_[0–11]_[a–b], tx_sync[a–d]) before txclk rises (2)
2.2
2
ns
th(TX)
Input hold (txin_[0–11]_[a–b], tx_sync[a–d]) after txclk rises (2)
1.1
ns
rises (2)
td(TX)
Data output delay (tx_sync_out_[0–5], tx_iflag, txout_[a–d]_[0–17]) after txclk
tOH(TX)
Data output hold (tx_sync_out_[0–5], tx_iflag, txout_[a–d]_[0–17]) after txclk rises (2)
FJCK
JTAG clock frequency (tck) (2)
tJCKL
JTAG clock low period (below VIL) (tck) (2)
(tck) (2)
ns
6.5
ns
40
MHz
1.5
ns
8
ns
tJCKH
JTAG clock high period (above VIH)
8
ns
tsu(J)
JTAG input (tdi or tms) setup before tck goes high (2)
2
ns
th(J)
JTAG input (tdi or tms) hold time after tck goes high (2)
9
ns
tck (2)
td(J)
JTAG output (tdo) delay from falling edge of
tsu(UPA)
Microprocessor address setup to falling edge of controls (2)
th(UPA)
Microprocessor address hold from rising edge of controls (2)
6
writes (2)
tsu(UPD)
Microprocessor data setup to rising edge of controls during
th(UPD)
Microprocessor data hold from rising edge of controls during writes (2)
th
Microprocessor data output hold from rising edge of controls (read)( (5)
td(UP)
Microprocessor data output delay from falling edge of controls (read) (2)
time (2)
tUPCKL
Microprocessor control low
tUPCKH
Microprocessor control high time (2)
(1)
(2)
(3)
(4)
(5)
ns
2.5
ns
2
ns
12
ns
2.6
ns
0
ns
36
ns
30
ns
8.4
ns
Timing is measured from the respective clock at VPAD/2 to input or output at VPAD/2. Output loading is a 50-Ω transmission line whose
delay is calibrated out.
Each part is tested at +90°C case temperature for the given specification. Lots are sample tested at –40°C.
Excluding tx_sync_out, tx_sync_out_[1–5].
Recommended practice.
Controlled by design and process and not directly tested. Verified on initial part evaluation.
Specifications
43
PACKAGE OPTION ADDENDUM
www.ti.com
2-Nov-2006
PACKAGING INFORMATION
Orderable Device
Status (1)
Package
Type
Package
Drawing
GC5318IZED
ACTIVE
BGA
ZED
Pins Package Eco Plan (2)
Qty
388
40
Pb-Free
(RoHS)
Lead/Ball Finish
SNAGCU
MSL Peak Temp (3)
Level-3-260C-168 HR
(1)
The marketing status values are defined as follows:
ACTIVE: Product device recommended for new designs.
LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect.
NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in
a new design.
PREVIEW: Device has been announced but is not in production. Samples may or may not be available.
OBSOLETE: TI has discontinued the production of the device.
(2)
Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check
http://www.ti.com/productcontent for the latest availability information and additional product content details.
TBD: The Pb-Free/Green conversion plan has not been defined.
Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements
for all 6 substances, including the requirement that lead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered
at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes.
Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and
package, or 2) lead-based die adhesive used between the die and leadframe. The component is otherwise considered Pb-Free (RoHS
compatible) as defined above.
Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame
retardants (Br or Sb do not exceed 0.1% by weight in homogeneous material)
(3)
MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder
temperature.
Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is
provided. TI bases its knowledge and belief on information provided by third parties, and makes no representation or warranty as to the
accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken and continues to take
reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on
incoming materials and chemicals. TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited
information may not be available for release.
In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI
to Customer on an annual basis.
Addendum-Page 1
IMPORTANT NOTICE
Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications,
enhancements, improvements, and other changes to its products and services at any time and to discontinue
any product or service without notice. Customers should obtain the latest relevant information before placing
orders and should verify that such information is current and complete. All products are sold subject to TI’s terms
and conditions of sale supplied at the time of order acknowledgment.
TI warrants performance of its hardware products to the specifications applicable at the time of sale in
accordance with TI’s standard warranty. Testing and other quality control techniques are used to the extent TI
deems necessary to support this warranty. Except where mandated by government requirements, testing of all
parameters of each product is not necessarily performed.
TI assumes no liability for applications assistance or customer product design. Customers are responsible for
their products and applications using TI components. To minimize the risks associated with customer products
and applications, customers should provide adequate design and operating safeguards.
TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right,
copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or process
in which TI products or services are used. Information published by TI regarding third-party products or services
does not constitute a license from TI to use such products or services or a warranty or endorsement thereof.
Use of such information may require a license from a third party under the patents or other intellectual property
of the third party, or a license from TI under the patents or other intellectual property of TI.
Reproduction of information in TI data books or data sheets is permissible only if reproduction is without
alteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproduction
of this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable for
such altered documentation.
Resale of TI products or services with statements different from or beyond the parameters stated by TI for that
product or service voids all express and any implied warranties for the associated TI product or service and
is an unfair and deceptive business practice. TI is not responsible or liable for any such statements.
Following are URLs where you can obtain information on other Texas Instruments products and application
solutions:
Products
Applications
Amplifiers
amplifier.ti.com
Audio
www.ti.com/audio
Data Converters
dataconverter.ti.com
Automotive
www.ti.com/automotive
DSP
dsp.ti.com
Broadband
www.ti.com/broadband
Interface
interface.ti.com
Digital Control
www.ti.com/digitalcontrol
Logic
logic.ti.com
Military
www.ti.com/military
Power Mgmt
power.ti.com
Optical Networking
www.ti.com/opticalnetwork
Microcontrollers
microcontroller.ti.com
Security
www.ti.com/security
Low Power Wireless www.ti.com/lpw
Mailing Address:
Telephony
www.ti.com/telephony
Video & Imaging
www.ti.com/video
Wireless
www.ti.com/wireless
Texas Instruments
Post Office Box 655303 Dallas, Texas 75265
Copyright  2006, Texas Instruments Incorporated