TI GC4016-PB

SLWS133A
GC4016
MULTI-STANDARD
QUAD DDC CHIP
DATA SHEET
REV 1.0
August 27, 2001
Graychip reserves the right to make changes in circuit design and/or specifications at any time
without notice. The user is cautioned to verify that datasheets are current before placing orders.
Information provided by Graychip is believed to be accurate and reliable. No responsibility is assumed by
Graychip for its use, nor for any infringement of patents or other rights of third parties which may arise
from its use. No license is granted by implication or otherwise under any patent rights of Graychip.
Ó
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
REVISION HISTORY
Revision
Date
Description
0.1
February 24, 2000
Original
0.2
April 10, 2000
Rewritten to reflect 2.5volt core, 3.3volt I/O power requirements,
Removed all references to TQ100 thin quad flat package,
All references to GC4017 are changed to GC4016,
Added Vcore and Vpin descriptions throughout,
Updated power consumption equations,
Simplified AC timing specifications,
Corrected Resampler Filter Specs,
Corrected typos, and miscellaneous errors.
0.3
April 20, 2000
Corrected Table 10: GC4016 Pin Out Locations, B14 and C14 VCORE and
VPAD swapped, other VPAD and GND pins added.
0.4
August 8, 2000
Minor Editing throughout
Page 12, Table 2, corrected +90 and -90 values
Added TAG_22 descriptions throughout
Added four separate frame strobe mode descriptions
Added resampler and ratio sync descriptions
Added one shot mode descriptions
Added parallel output timing diagram
Revised initialization procedures
1.0
August 27, 2001
Minor editing throughout
Electrical and Timing Specifications in Section 6 updated to match production
test
References to LVDS levels changed to differential inputs.
Diagnostics, GC4016 Configuration Generator, IS95 and UMTS examples
added
 GRAYCHIP,INC.
-i-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
CONTACTING GRAYCHIP
CORPORATE OFFICES:
GRAYCHIP, Inc.
2185 Park Blvd.
Palo Alto, CA 94306
PHONE:
(650) 323-2955
FAX:
(650) 323-0206
WEB PAGE:
www.graychip.com
E-MAIL:
[email protected]
[email protected]
 GRAYCHIP,INC.
- ii -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Table of Contents
1.0
KEY FEATURES ......................................................................................................................... 1
2.0
BLOCK DIAGRAM ...................................................................................................................... 1
3.0
FUNCTIONAL DESCRIPTION .................................................................................................... 2
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
3.10
3.11
3.12
3.13
3.14
3.15
3.16
4.0
CONTROL INTERFACE ............................................................................................................................................ 2
INPUT FORMAT ........................................................................................................................................................ 4
THE DOWN CONVERTERS ..................................................................................................................................... 4
MULTICHANNEL MODES....................................................................................................................................... 10
RESAMPLER........................................................................................................................................................... 12
OVERALL GAIN ...................................................................................................................................................... 15
PEAK COUNTER .................................................................................................................................................... 15
OUTPUT MODES .................................................................................................................................................... 16
CLOCKING .............................................................................................................................................................. 22
POWER DOWN MODES......................................................................................................................................... 23
SYNCHRONIZATION .............................................................................................................................................. 24
INITIALIZATION ...................................................................................................................................................... 25
DATA LATENCY...................................................................................................................................................... 25
DIAGNOSTICS ........................................................................................................................................................ 25
JTAG........................................................................................................................................................................ 25
MASK REVISION REGISTER ................................................................................................................................. 26
PACKAGING ............................................................................................................................. 27
4.1
5.0
GC4016-PB 160 Ball Plastic Ball Grid Array (PBGA) .............................................................................................. 27
CONTROL REGISTERS ........................................................................................................... 31
5.1
5.2
5.3
5.4
5.5
5.6
5.7
5.8
5.9
5.10
5.11
6.0
GLOBAL CONTROLS ............................................................................................................................................. 31
PAGED REGISTERS .............................................................................................................................................. 34
CFIR COEFFICIENT PAGES .................................................................................................................................. 35
PFIR COEFFICIENT PAGES .................................................................................................................................. 35
CHANNEL FREQUENCY PAGES........................................................................................................................... 36
CHANNEL CONTROL PAGES................................................................................................................................ 37
RESAMPLER COEFFICIENT PAGES .................................................................................................................... 42
RESAMPLER CONTROL PAGE ............................................................................................................................. 43
RESAMPLER RATIO PAGE) .................................................................................................................................. 45
CHANNEL OUTPUT PAGES .................................................................................................................................. 45
OUTPUT CONTROL PAGE .................................................................................................................................... 46
SPECIFICATIONS..................................................................................................................... 53
6.1
6.2
6.3
6.4
6.5
6.6
7.0
ABSOLUTE MAXIMUM RATINGS .......................................................................................................................... 53
RECOMMENDED OPERATING CONDITIONS ...................................................................................................... 53
THERMAL CHARACTERISTICS............................................................................................................................. 53
POWER CONSUMPTION ....................................................................................................................................... 54
DC CHARACTERISTICS......................................................................................................................................... 54
AC CHARACTERISTICS ......................................................................................................................................... 55
APPLICATION NOTES ............................................................................................................. 56
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
7.11
7.12
7.13
7.14
POWER AND GROUND CONNECTIONS .............................................................................................................. 56
STATIC SENSITIVE DEVICE .................................................................................................................................. 56
MOISTURE SENSITIVE PACKAGE........................................................................................................................ 56
THERMAL MANAGEMENT ..................................................................................................................................... 56
EXAMPLE CFIR FILTER SETS............................................................................................................................... 57
EXAMPLE PFIR FILTER SETS ............................................................................................................................... 58
EXAMPLE RESAMPLER CONFIGURATIONS ....................................................................................................... 60
GC4016 CONFIGURATION GENERATOR............................................................................................................. 61
EXAMPLE GSM APPLICATION .............................................................................................................................. 62
EXAMPLE IS-136 DAMPS APPLICATION.............................................................................................................. 64
IS-95 NB-CDMA APPLICATION.............................................................................................................................. 66
UMTS WB-CDMA APPLICATION ........................................................................................................................... 68
DIAGNOSTICS ........................................................................................................................................................ 71
OUTPUT TEST CONFIGURATION......................................................................................................................... 73
 GRAYCHIP,INC.
- iii -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
LIST OF FIGURES
Figure 1:
Figure 2:
Figure 3:
Figure 4:
Figure 5:
Figure 6:
Figure 7:
Figure 8:
Figure 9:
Figure 10:
Figure 11:
Figure 12:
Figure 13:
Figure 14:
Figure 15:
Figure 16:
Figure 17:
Figure 18:
Figure 19:
Figure 20:
Figure 21:
Figure 22:
Figure 23:
GC4016 Block Diagram ......................................................................................................................................................... 1
Normal Control I/O Timing ..................................................................................................................................................... 3
Edge Write Control Timing..................................................................................................................................................... 3
The Down Converter Channel ............................................................................................................................................... 5
Zero Pad Synchronization...................................................................................................................................................... 6
NCO Circuit............................................................................................................................................................................ 6
Example NCO Spurs.............................................................................................................................................................. 7
NCO Peak Spur Plot .............................................................................................................................................................. 7
Five Stage CIC Decimate by N Filter .....................................................................................................................................8
Typical CFIR Specifications ................................................................................................................................................... 9
Typical PFIR Specifications ................................................................................................................................................... 9
Resampler Channel Block Diagram..................................................................................................................................... 12
The Resampler’s Spectral Response .................................................................................................................................. 13
Wide Word Microprocessor Port .......................................................................................................................................... 19
Serial Output Formats.......................................................................................................................................................... 21
Parallel Mode Timing ........................................................................................................................................................... 23
GC4016 160 Pin Plastic Ball Grid Array .............................................................................................................................. 27
GC4016 Pin Assignments.................................................................................................................................................... 28
Overall Spectral Responses of Example Filters................................................................................................................... 59
Frequency Response for the Example GSM Application ..................................................................................................... 62
Frequency Response for the Example DAMPS Application ................................................................................................ 64
Frequency Response for the Example IS95 Application...................................................................................................... 66
Frequency Response for the Example UMTS Application................................................................................................... 69
Table 1:
Table 2:
Table 3:
Table 4:
Table 5:
Table 6:
Table 7:
Table 8:
Table 9:
Table 10:
Table 11:
Table 12:
Table 13:
Table 14:
Table 15:
Table 16:
Table 17:
Table 18:
Table 19:
Table 20:
Table 21:
Table 22:
Table 23:
Table 24:
Table 25:
Table 26:
Table 27:
Table 28:
Table 29:
Table 30:
Table 31:
Table 32:
Table 33:
Table 34:
Table 35:
Table 36:
Table 37:
Table 38:
Table 39:
Table 40:
Input Mode Controls...............................................................................................................................................................4
Multichannel mode settings ................................................................................................................................................. 12
Output Mode Controls.......................................................................................................................................................... 17
Channel Output Map, Synchronous Four Channel Mode .................................................................................................... 18
Channel Output Map, Synchronous Two Channel Mode..................................................................................................... 18
Channel Output Map, Single Channel Mode ....................................................................................................................... 18
Sync Modes ......................................................................................................................................................................... 24
Recommended Sync Settings.............................................................................................................................................. 24
Mask Revisions.................................................................................................................................................................... 26
GC4016 Pin Out Locations Top View .................................................................................................................................. 29
Global Control Registers...................................................................................................................................................... 31
Page Assignments ............................................................................................................................................................... 34
CFIR Coefficient Pages ....................................................................................................................................................... 35
PFIR Coefficient Pages........................................................................................................................................................ 35
Channel Control Registers................................................................................................................................................... 37
Resampler Coefficient Pages (Single filter mode) ............................................................................................................... 42
Resampler Control Registers............................................................................................................................................... 43
Resampler Ratio Page......................................................................................................................................................... 45
Output Control Registers ..................................................................................................................................................... 46
Absolute Maximum Ratings ................................................................................................................................................. 53
Recommended Operating Conditions.................................................................................................................................. 53
Thermal Data ....................................................................................................................................................................... 53
DC Operating Conditions ..................................................................................................................................................... 54
AC Characteristics ............................................................................................................................................................... 55
Example CFIR Filters........................................................................................................................................................... 57
Default PFIR Filters.............................................................................................................................................................. 58
Resampler Bypass Mode..................................................................................................................................................... 60
Resampler Modes................................................................................................................................................................ 60
Desired GSM Specifications ................................................................................................................................................ 62
Example GSM Configuration ............................................................................................................................................... 63
Desired DAMPS Specifications............................................................................................................................................ 64
Example DAMPS Configuration........................................................................................................................................... 65
Desired IS95 NB-CDMA Specifications ............................................................................................................................... 66
Example IS95 NB-CDMA Configuration .............................................................................................................................. 68
Desired UMTS Specifications .............................................................................................................................................. 68
Example UMTS Configuration ............................................................................................................................................. 70
Diagnostic Test 1 ................................................................................................................................................................. 71
Diagnostic Test 2 ................................................................................................................................................................. 71
Diagnostic Test 3 ................................................................................................................................................................. 72
Diagnostic Test 4 ................................................................................................................................................................. 73
LIST OF TABLES
 GRAYCHIP,INC.
- iv -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
KEY FEATURES
BLOCK DIAGRAM
B IN
(14 BITS)
A IN
(14 BITS)
CROSSBAR SWITCH
DIAGNOSTICS
CONTROL INTERFACE
21 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(CFIR)
A[0:4]
FILTER
21 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(CFIR)
FILTER
21 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(CFIR)
GENERATOR
NCO
SINE/COSINE
NCO
GENERATOR
DECIMATE
BY 4 TO 4K
DECIMATE
BY 4 TO 4K
FILTER
21 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(CFIR)
PHASE
OFFSET
WR
DECIMATE
BY 4 TO 4K
TUNING
FREQUENCY
FILTER
TUNING
FREQUENCY
DECIMATE
BY 4 TO 4K
CE
RD
GENERATOR
NCO
(16 bit input and 12 bit floating point modes use the DIN pins for the LSB’s of inputs AIN, BIN, and CIN)
SINE/COSINE
CLOCK DOUBLING
AND
DISTRIBUTION CIRCUIT
PHASE
OFFSET
CK
DIN
(14 BITS)
CIN
(14 BITS)
SINE/COSINE
2.0
PHASE
OFFSET
•
•
•
•
•
•
TUNING
FREQUENCY
•
•
•
•
•
NCO
•
•
•
•
GENERATOR
•
•
•
•
•
SINE/COSINE
•
Input rates up to 100 MSPS
Four independent digital down convert (DDC)
channels
Single channel GC4011 and Dual channel
GC4012 derivatives are available
Independent decimation and resampling
Independent tuning, phase and gain controls
Input Crossbar Switch for:
Four 14 bit Inputs, or
Three 16 bit Inputs, or
Three 12 bit + 3 bit exponent inputs, or
Two 14 bit differential inputs.
Decimation factors of
32 to 16,384 in each channel
16 to 32 by combining two channels
8 to 16 by combining four channels
Zero padding for lower decimation factors
Resampler for arbitrary decimation factors
Peak detection counters for AGC loop controls
Outputs can be either:
Bit serial, or
Nibble serial (link port), or
Parallel port, or
Memory mapped registers
12, 16, 20, or 24 bit output samples
0.02 Hz tuning resolution
>100 dB far band rejection
>115 dB spur free dynamic range
User programmable 21 tap and 63 tap decimate by
two filters, independent per channel.
Nyquist filtering for QPSK or QAM symbol data
Resampler provides additional filtering and allows
arbitrary input/output rate selections
Microprocessor interface for control
Built in diagnostics
Application examples:
Four 4X oversampled GSM, DAMPS, or IS95
CDMA carriers,
Two 8X oversampled IS95 CDMA carriers, or
Two 2X oversampled 3.84MB UMTS carriers, or
One 4X oversampled 3.84MB UMTS carrier
Core power consumption at 80 MHz, 2.5 volts:
100 mW per DAMPS channel
115 mW per GSM channel
115 mW per IS95 channel
620 mW per 3.84MB UMTS channel
Industrial temperature range (-40C to +85C)
GC4016-PB 160 ball PBGA
(15mm by 15mm) package
3.3volt I/O voltage, 2.5volt core voltage
JTAG Boundary Scan
PHASE
OFFSET
•
•
•
•
•
•
TUNING
FREQUENCY
1.0
DATA SHEET REV 1.0
C[0:7]
TCK
TMS
TDI
TDO
63 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(PFIR)
63 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(PFIR)
63 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(PFIR)
63 TAP
PROGRAMMABLE
DECIMATE BY 2
FILTER
(PFIR)
JTAG
BOUNDARY SCAN TEST
LOGIC
RESAMPLER AND GAIN BLOCK
COMBINES CHANNELS, FILTERS, RESAMPLES AND ADJUSTS GAIN
ACKNOWLEDGE
(LINK MODE)
(FILTERING, RESAMPLING AND GAIN ADJUSTMENTS ARE PERFORMED INDEPENDENTLY ON EACH CHANNEL)
24 BIT PARALLEL OUTPUT
A B C D
(BIT SERIAL OUTPUTS)
Figure 1. GC4016 Block Diagram
 GRAYCHIP,INC.
-1-
READY
SO
OUTPUT FORMAT
FRAME SYNCS
SIB
SYNC COUNTER
AND
DIAGNOSTIC TEST
GENERATOR
BIT-CLOCK
SIA
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
3.0
DATA SHEET REV 1.0
FUNCTIONAL DESCRIPTION
3.1
The GC4016 quad receiver chip contain four identical
down-conversion circuits. Each downconvert circuit accepts
a real sample rate up to 100 MHz, down converts a selected
carrier frequency to zero, decimates the signal rate by a
programmable factor ranging from 32 to 16,384 and then
resamples the channel to adjust the sample rate up or down
by an arbitrary factor. In the real output mode the output
sample rate is doubled and the signal is output as a real
signal centered at Fout/4. The channels may be combined to
The chip is configured by writing control information into
control registers within the chip. The control registers are
grouped into 8 global registers and 128 pages of registers,
each page containing up to 16 registers. The global registers
are accessed as addresses 0 through 7. Address 2 is the
page register which selects which page is accessed by
addresses 16 through 31. The contents of these control
registers and how to use them are described in Section 5.
The registers are written to or read from using the
C[0:7], A[0:4], CE, RD and WR pins. Each control register
has been assigned a unique address within the chip. This
interface is designed to allow the GC4016 chip to appear to
an external processor as a memory mapped peripheral (the
pin RD is equivalent to a memory chip’s OE pin).
produce wider band and/or oversampled outputs or to
process complex input data. The chip outputs the
down-converted signals in any one of several formats
(microprocessor, four serial lines, one TDM serial line,
nibble, LINK, or 24 bit parallel port. The chip contains two
user programmable output filters per path which can be used
to arbitrarily shape the received data’s spectrum. These
filters can be used as Nyquist receive filters for digital data
transmission. The chip also contains a resampling filter to
provide additional filtering and to allow the user complete
flexibility in the selection of input and output sample rates.
An external processor (a microprocessor, computer, or
DSP chip) can write into a register by setting A[0:4] to the
desired register address, selecting the chip using the CE pin,
setting C[0:7] to the desired value and then pulsing WR low.
The data will be written into the selected register when both
WR and CE are low and will be held when either signal goes
high. An alternate “edge write” mode can be used to strobe
the data into the selected register when either WR or CE
goes high. This is useful for processors that do not guarantee
valid data when the write strobe goes low, but guarantee that
the data will be stable before the write strobe goes high. The
edge write mode is necessary for these processors, as some
control registers (such as most sync or reset registers) are
sensitive to transient values on the C[0:7] data bus.
Two downconverter paths can be merged to be used as
a single complex input down-conversion circuit. Two paths
may also be combined to support wider band output rates or
oversampled outputs. Four paths may be combined to
support both wider band output and oversampling.
The downconverters are designed to maintain over 115
dB of spur free dynamic range and over 100 dB of out of band
rejection. A five stage CIC and 20 bit internal datapaths
support this high dynamic range signal processing
requirement. Each downconvert circuit accepts 16 bit inputs
and produces 24 bit outputs (can be rounded back to 12, 16,
or 20 bits). The frequencies and phase offsets of the four
sine/cosine sequence generators can be independently
specified, as can the decimation and filter parameters of
each circuit.
To read from a control register the processor must set
A[0:4] to the desired address, select the chip with the CE pin,
and then set RD low. The chip will then drive C[0:7] with the
contents of the selected register. After the processor has
read the value from C[0:7] it should set RD and CE high. The
C[0:7] pins are turned off (high impedance) whenever CE or
RD are high or when WR is low. The chip will only drive these
pins when both CE and RD are low and WR is high.
On chip diagnostic circuits are provided to simplify
system debug and maintenance.
One can also ground the RD pin and use the WR pin as
a read/write direction control and use the CE pin as a control
I/O strobe. Figure 2 shows timing diagrams illustrating both
I/O modes.
The chip receives configuration and control information
over a microprocessor compatible bus consisting of an 8 bit
data I/O port, a 5 bit address port, a chip enable strobe, a
read strobe and a write strobe. The chip’s control registers (8
bits each) are memory mapped into the 5 bit address space
of the control port.
The edge write mode, enabled by the EDGE_WRITE
control bit in register 0, allows for rising edge write cycles. In
this mode the C[0:7] data only needs to be stable for a setup
time before the rising edge of the write strobe, and held for a
small hold time afterwards. This mode is appropriate for
Sections 7.9 through 7.12 describe how to use the chip
for GSM, D-AMPS, CDMA and UMTS applications, including
control register values and filter coefficients.
 GRAYCHIP,INC.
CONTROL INTERFACE
-2-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
CE
WR
DATA SHEET REV 1.0
tREC
tCSU
RD
tREC
tCSU
A[0-4]
tCDLY
C[0-7]
tCZ
READ CYCLE- NORMAL MODE
CE
WR
tREC
tCSPW
tCSU
RD
tREC
tCSU
A[0-4]
tCHD
C[0-7]
WRITE CYCLE- NORMAL MODE
CE
tREC
WR
tREC
tCSU
A[0-4]
tCDLY
C[0-7]
tCZ
READ CYCLE- RD HELD LOW
CE
WR
tCSPW
tREC
tCSU
A[0-4]
tREC
tCHD
C[0-7]
WRITE CYCLE- RD HELD LOW
Figure 2. Normal Control I/O Timing
chip’s expansion bus. Address pin A3 is grounded in this
mode and the LSB of the page register (address 2) is used in
its place. The four bit mode is turned on using the
4_BIT_ADDRESS control bit in address 4.
processors that do not provide stable data before the start of
the write pulse. Figure 3 shows the timing for this mode.
The setup, hold and pulse width requirements for control
read or write operations are given in Section 6.0.
The chip also operates in a four bit address mode which
is intended to be used with the 4 address bit TI320C6X DSP
CE
WR
RD
tREC
tCSPW
tCSU
tREC
tCSU
A[0-4]
tCHD
C[0-7]
EDGE WRITE MODE
tEWCSU
Figure 3. Edge Write Control Timing
 GRAYCHIP,INC.
-3-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
3.2
DATA SHEET REV 1.0
3.3
INPUT FORMAT
THE DOWN CONVERTERS
Each down converter contains an NCO and a mixer to
quadrature down convert the signal to baseband, followed by
The chip accept five input formats:
(1) Four input ports of 14 bit data,
a 5 stage Cascade Integrate Comb (CIC)1 filter and two
stages of decimate by two filtering to isolate the desired
signal. The signal is then sent to a resampler which can
increase or decrease the final output sample rate to match
the post-processing requirements for baud rate sampling or
oversampling.
(2) Three input ports of 16 bit data,
(3) Two input ports of 14 bit low voltage differential
data, (Contact Graychip for more information)
(4) Three ports of 12 bit with 3 bit exponent floating
point data, and
(5) Three ports of multiplexed, dual channel, 12 bit
with 3 bit exponent floating point data
A block diagram of the channel is shown in Figure 4.
The INPUT FORMAT circuit converts the selected input
data into the 19 bit format described in Section 3.2. The
ZERO PAD function allows the user to clock the chip at a
higher rate than the input sample rate.
The 12, 14 and 16 bit input words are in a two’s
complement format. The MSB_POL control bit in address 27
of the channel control pages can be used to convert offset
binary data to the desired two’s complement format. The 3 bit
exponent in the floating point format is an unsigned integer
ranging from 0 to 7. All inputs are converted to the internal 19
bit format at the input to each channel. The 14 and 16 bit
input words are put into the upper 14 and 16 bits respectfully
of the 19 bit word. The unused LSBs are cleared.
The NCO/Mixer circuit tunes the desired center
frequency down to DC where it is low pass filtered by the
CIC, CFIR, PFIR and Resampler filters.
The CIC filter reduces the sample rate by a
programmable factor ranging from 8 to 4,096. The CIC
outputs are followed by a coarse gain stage and then
followed by two stages of decimate by 2 filtering. The coarse
gain circuit allows the user to boost the gain of weak signals
up to 42 dB in 6 dB steps. The first stage of the two stage filter
is a 21 tap decimate by 2 filter (CFIR) with user
programmable tap weights. The 21 tap symmetrical lowpass
filter is downloaded into the chip as 12 words, 16 bits each.
This filter is typically programmed to decimate by two, while
The 12 bit floating point word is shifted down and sign
extended by the amount specified by the 3 bit exponent and
then put into the MSBs of 19 bit word.
A crossbar switch allows the user to route any input
source to any downconverter channel.
Table 1 shows the suggested control register settings
required for each input mode. See Section 5.6 for detailed
descriptions of each control setting
1. Hogenhauer, An Economical Class of Digital Filters for Decimation and
Interpolation, IEEE transactions on ASSP, April 1981.
Table 1: Input Mode Controls
16 bit mode
14 bit
differential
mode
12 + 3 bit
exponent
mode
Multiplexed
12 + 3 bit
exponent
mode
4 to 7
4 to 7
4 to 7
unused
unused
16
1
1
1
0
0
23
1
1
1
0
0
Control
Address in
Channel Control
Pages 7, 15, 23
and 31
14 bit mode
SHIFT1
16
USE_SHIFT
MIX20B1
INPUT_SEL
3
SEL_AB
27
0,1,2,3
0,1,2
4,5
0,1,2
0,1,2
27
0
0
0
0
0 or 1
INPUT_MODE
27
0
1
0
2
2
DIFF_IN
4 (Global Page)
0
0
1
0
0
NOTES: 1. SHIFT is normally left at 4, see Section 3.3.3 for details.
2. For decimations over 3104, MIX20B should be 0 to allow SHIFT to be less than 4, see Section 3.3.3.
3. INPUT_SEL is used to select the input port used by each channel, 0=port A, 1=portB, 2=port C, 3=portD.
 GRAYCHIP,INC.
-4-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
compensating for the droop in the CIC filter’s passband. The
second stage is a 63 tap decimate by 2 filter (PFIR) with user
programmable tap weights. The user can customize the
channel’s spectral response using the PFIR filter. A typical
use of the PFIR is to perform matched (root-raised cosine)
filtering. The 63 tap symmetrical filter is downloaded into the
chip as 32 words, 16 bits each. Both filters must be
programmed as there are no default filter coefficient sets.
bit word (6 integer bits and 26 fractional bits). The time
resolution of the new sampling points is user programmable
down to 1/64th of the PFIR output’s sample period. If R is a
multiple of the selected time resolution, then the resampling
process will have no resampling time jitter. If R is not a
multiple of the time resolution, then the output will be
generated at the time instant closest to the desired output
time.
The CIC, CFIR, and PFIR filters can all be configured in
a SPLITIQ mode where the channel only processes the real
portion of the mixer output. In this case the minimum CIC
decimation is 4. By using two channels in parallel one can
process twice the output bandwidth by processing the real
portion of the mixer output in one channel and the imaginary
portion in a second channel (hence the name “split I/Q”). One
can also offset the decimation timing between channels so
that two or even four channels can be used to generate
oversampled data. Using this capability each GC4016 chip
can downconvert wideband signals such as 4X oversampled
5.12 Msymbol/sec MCNS data, or 4X oversampled 5 MHz
WB-CDMA (UMTS) data, or two 8X oversampled NB-CDMA
IS95 signals.
The resampler is typically used to either generate
oversampled output data (for instance 4x oversampled
QPSK or GMSK data), or to alter the sampler rate to match
a multiple of the baud rate of a QPSK or GMSK signal. The
resampling operation allows the user to design a system
where the ADC sample rate does not need to be equal to an
integer multiple of the baud rate of the desired output signal.
The resampler can also be used as a final filtering stage.
A final shift circuit allows the user to shift the data by up
to 15 bits to properly place it in the output word. The output
word may be rounded to 12,16, 20, or 24 bits.
The data may be output from the chip via a
microprocessor interface, serial pins, or using a parallel port.
The parallel port is particularly valuable for wideband output
data.
The PFIR will also, if desired, convert the complex
output data to real. The complex to real conversion also
doubles the output sample rate so that the PFIR does not
decimate in this mode. This mode is useful when one wants
to output real data.
3.3.1 Zero Padding
20
20
FINAL SHIFT
CHANNELS)
RESAMPLER
RES TAPS
RATIO
FINE GAIN
PFIR TAPS
PFIR FILTER
DECIMATE
BY 2
CFIR FILTER
DECIMATE
BY 2
COARSE
GAIN
24
(SHARED BETWEEN
20
CIC FILTER
DECIMATE BY
8 TO 4K
19
INPUT FORMAT
AND
ZERO PAD
IN
12-16 bits
N
COARSE
The PFIR filter is followed by a resampler. The
resampler filters the PFIR output to generate new sample
points in between the PFIR output samples. The resampler
is programmed to generate new output samples spaced in
time by “R” times the PFIR output’s sample period. The value
R is called the resampling ratio. If R is less than one, the
resampler will interpolate. If R is greater than one, the
resampler will decimate. The value of R is specified as a 32
CFIR TAPS
The input samples are normally clocked into the chip at
the clock rate, i.e., the input sample rate is equal to the clock
rate. Input rates lower than the clock rate can be accepted by
using the zero pad mode. When enabled by setting the
ZPAD_EN bit in address 19 of the channel control pages, the
zero pad mode will insert “NZERO” zeroes between each
input sample, where NZERO ranges from 0 to 15, allowing
input data rates down to 1/16th the clock rate. NZERO is set
in address 19
24
I
TO
OUTPUT
CIRCUIT
Q
12,16, 20 or 24 bits
20
TUNING
FREQUENCY
PHASE
OFFSET
NCO
Figure 4. The Down Converter Channel
 GRAYCHIP,INC.
-5-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
CK
Sample taken
SIB
Input
Data is sampled every
(NZERO+1) clocks thereafter
3+NZERO CLOCKS
Figure 5. Zero Pad Synchronization
The tuning frequency is set to FREQ according to the
Zero padding lowers the effective decimation ratio. For
example, the minimum complex output decimation using a
single channel is normally 32. If the input data rate is 5 MSPS
and the system can clock the chip at 40 MHz, then the zero
pad function can be used to insert seven zeros between each
sample, padding the 5 MSPS input data rate up by a factor of
eight to 40 MSPS. The minimum decimation of 32 from the
40MHz rate results in an output rate of 1.25 MSPS, which is
an effective decimation of 4 relative to the original 5 MSPS
data.
formula FREQ = 232F/FCK, where F is the desired tuning
frequency and FCK is the chip’s clock rate. The 16 bit phase
offset setting is PHASE = 216P/2π, where P is the desired
phase in radians ranging between 0 and 2π.
Note that a positive tuning frequency should be used to
downconvert the signal. A negative tuning frequency can be
used to upconvert the negative (spectrally flipped) image of
the desired signal. FREQ and PHASE are set in addresses
16 through 21 of each channel frequency pages.
A sync signal is used to synchronize the zero padding.
The ZPAD_SYNC control in address 19 selects the source of
the sync signal. The sync signal can be used to identify when
the chip will sample the input data. Figure 5 shows the timing
when SIB (ZPAD_SYNC=3) is selected as the sync source.
The NCO’s frequency, phase and accumulator can be
initialized and synchronized with other channels using the
FREQ_SYNC, PHASE_SYNC, and NCO_SYNC controls in
addresses 17 and 18 of the channel control pages. The
FREQ_SYNC and PHASE_SYNC controls determine when
new frequency and phase settings become active. Normally
these are set to “always” so that they take effect immediately,
but can be used to synchronize frequency hopping or beam
forming systems. The NCO_SYNC control is usually set to
never, but can be used to synchronize the LOs of multiple
channels.
Zero padding can be used to synchronize extracting
data from a TDM bus. By adjusting the timing of SIB as
shown in Figure 5, the user can choose which sample to take
from the TDM bus.
The zero pad function has a gain equal to:
ZPAD_GAIN =1/(1+NZERO).
The NCO’s spur level is reduced to below -113 dB
through the use of phase dithering. The spectrums in Figure
7 show the NCO spurs for a worst case tuning frequency
before and after dithering has been turned on. Notice that the
spur level decreases from -105 dB to -116 dB. Dithering is
turned on or off using the DITHER_SYNC control in address
18.
3.3.2 The Numerically Controlled Oscillator
(NCO)
The tuning frequency of each down converter is
specified as a 32 bit word and the phase offset is specified as
a 16 bit word. The NCOs can be synchronized with NCOs on
other chips. This allows multiple down converter outputs to
be coherently combined, each with a unique phase and
amplitude. A block diagram of the NCO circuit is shown in
Figure 6.
The worst case NCO spurs at -113 to -116dB, such as
the one shown in Figure 7(b), are due to a few frequencies
that are related to the sampling frequency by multiples of
FCK/96 and FCK/124. In these cases the rounding errors in
DITHER
PHASE GENERATOR
32 BITS
32 BITS
16 BITS
23 MSBs
5 BITS
18 MSBs
FREQ
SINE/COSINE
LOOKUP
TABLE
20 BITS
SINE/COSINE
OUT
Figure 6. NCO Circuit
 GRAYCHIP,INC.
-6-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
FREQ=5/24 F S
FREQ=5/24 F S
-105 dB
-116 dB
a) Worst case spectrum without dither
b) Spectrum with dither (tuned to same frequency)
Figure 7. Example NCO Spurs
-107 dB
-121 dB
a) Plot without dither or phase initialization
b) Plot with dither and phase initialization
Figure 8. NCO Peak Spur Plot
achieve a CIC gain of less than or equal to unity when the
CIC decimation is greater than 3104. See Table 1.
the sine/cosine lookup table repeat in a regular fashion,
thereby concentrating the error power into a single
frequency, rather than spreading it across the spectrum.
These worst case spurs can be eliminated by selecting an
initial phase that minimizes the errors or by changing the
tuning frequency by a small amount (50 Hz). Setting the
initial phase to 4 for multiples of FCK/96 or FCK/124 (and to 0
3.3.3 Five Stage CIC Filter
The mixer outputs are decimated by a factor of N in a
five stage CIC filter, where N is any integer between 8 and
4096 (between 4 and 2048 for SPLITIQ mode). The value of
N is programmed independently for each channel in
addresses 21 and 22 of each channel control page. The
programmable decimation allows the chip’s usable output
bandwidth to range from less than 4 kHz to over 3 MHz when
the input rate is 100 MHz. Wider output bandwidths are
obtainable by using multiple channels (see Section 3.4). A
block diagram of the CIC filter is shown in Figure 9.
for other frequencies) will result in spurs below -115 for all
frequencies.
Figure 8 shows the maximum spur levels as the tuning
frequency is scanned over a portion of the frequency range
with the peak hold function of the spectrum analyzer turned
on. Notice that the peak spur level is -107 dB before dithering
and is -121 dB after dithering has been turned on and the
phase initialization described above has been used.
The output of the mixer may be rounded to 16 or 20 bits.
Twenty bit rounding (MIX20B is set in address 23) is normally
used with the 14 or 16 bit input modes. Sixteen bit rounding
(MIX20B disabled) is normally used with the floating point, 12
bit input modes. Sixteen bit rounding is also required to
 GRAYCHIP,INC.
-7-
August 27, 2001
This document contains information which may be changed at any time without notice
CIC_SCALE
DATA 20 BITS
IN
DATA SHEET REV 1.0
DECIMATE
BY FACTOR
OF N
GC4016 MULTI-STANDARD QUAD DDC CHIP
CLOCKED AT FULL RATE
UPPER
24 BITS
24 BITS
DATA
OUT
CLOCKED AT 1/N RATE
Figure 9. Five Stage CIC Decimate by N Filter
3.3.5 The First Decimate By Two Filter
(CFIR)
The CIC filter has a gain equal to N5 which must be
compensated for in the “CIC_SCALE” circuit shown in Figure
9. The CIC_SCALE circuit has a gain equal to
The CIC/Coarse gain outputs are filtered by two stages
of filtering. The first stage is a 21 tap decimate by 2 filter with
programmable 16 bit coefficients. Since this filter decimates
by two, a stopband must be created in that portion of the
spectrum that would alias into the signal of interest. This filter
has very lax transition band specifications so 21 taps is
sufficient to both provide the required anti-aliasing stopband,
and to provide compensation for the droop in the CIC filter’s
passband. The CFIR is also used, in some cases, to provide
additional stopband rejection for the second stage PFIR filter.
Figure 10 illustrates the passband and stopband
requirements of the filter. FCFIR is the input sample rate to the
2(SHIFT+SCALE+6*BIG_SCALE-62), where SCALE ranges from 0
to 5 and BIG_SCALE ranges from 0 to 7. The range of SHIFT
is 4-7 if MIX20B is enabled and is 0-7 if MIX20B is disabled.
The overall gain of the CIC circuit is equal to:
5
CIC_GAIN = N 2 ( SHIFT +SCALE
+ 6 × BIG_SCALE – 62 )
The user must select values for SHIFT, SCALE and
BIG_SCALE (addresses 16 and 23 of each channel control
page) such that CIC_GAIN (including ZPAD_GAIN if
blanking is used) is less than one, i.e., SHIFT, SCALE and
BIG_SCALE must be selected such that:
CFIR filter. FCFIR/4 is the output sample rate of the channel
( SHIFT + SCALE + 6 × BIG_SCALE ) ≤
before resampling.
( 62 – 5log2 N + log2 (NZERO+1) )
The user downloaded filter coefficients are 16 bit 2’s
complement numbers. Unity gain will be achieved through
the filter if the sum of the 21 coefficients is equal to 65536. If
the sum is not 65536, then CFIR will introduce a gain equal
Overflows due to improper gain settings will go
undetected if this relationship is violated. For example, if N is
equal to 8 and SHIFT=4, then this restriction means that
BIG_SCALE and SCALE should be less than or equal to 7
and 1 respectively. The SHIFT, BIG_SCALE and SCALE
settings are independent for each channel. See Section 3.7
for a description of the channel’s overall gain.
to:
The gain of each channel can be boosted up to 42 dB by
shifting the output of the CIC filter up by 0 to 7 bits prior to
rounding it to 20 bits. The coarse gain is:
COARSE
The 21 coefficients are identified as coefficients h0
through h20, where h10 is the center tap. The coefficients are
, where COARSE ranges from
assumed to be symmetric, so only the first 11 coefficients (h0
0 to 7. COARSE is set in address 25 of each channel control
page. Overflows in the coarse gain circuit are saturated to
plus or minus full scale. The coarse gain is used to increase
the gain of an individual signal after the input bandwidth of
the downconverter has been reduced by a factor of N in the
CIC filter. If the signal power across the input bandwidth is
relatively flat, as is the case in most frequency division
multiplexed (FDM) systems, then one would want to boost
the signal power out of the CIC filter by a factor of
COARSE_GAIN =
, where CFIR_SUM is the sum of
the 21 coefficients. Coefficient sets for a variety of standards
in cellular and cable modem applications are given in Section
7. The output of CFIR is rounded to 20 bits (using the
round-to-even algorithm). Overflows are detected and hard
limited. Overflows can be directed to the channel overflow
detection block.
3.3.4 Coarse Channel Gain
COARSE_GAIN = 2
CFIR_SUM
CFIR_GAIN = ----------------------------65536
through h10) are loaded into the chip. A non-symmetric mode
(NO_SYM_CFIR in address 25) allows the user to download
an 11 tap non-symmetric filter as taps h0 through h10. The
newest sample is multiplied by h0 and the oldest is multiplied
by h10. NOTE: Filters normally multiply h0 by the oldest data,
hence one may wish to reverse the tap order in the
non-symmetric mode.
CFIR has a programmable delay of one CFIR input
sample. This delay is used in a multichannel mode to alter
N . Each channel can be given its own
coarse gain setting. See Section 3.7 for a description of the
channel’s overall gain.
 GRAYCHIP,INC.
-8-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Signal of Interest Passband (Typically 0.05 to 0.125 FCFIR)
0 dB
Very wide transition band
Power
Stopband (Typically starts at 0.375 F CFIR)
-100 dB
0
FCFIR/8
FCFIR/4
FCFIR/2
Frequency
Figure 10. Typical CFIR Specifications
the filter if the sum of the 63 coefficients is equal to 65536. If
the sum is not 65536, then PFIR will introduce a gain equal
the decimate by two phasing in order to generate
oversampled or wider bandwidth output data. This delay is
independently programmable for each channel and is also
independent for I and Q (QDLY_CFIR and IDLY_CFIR in
address 25). A special mode (SPLITIQ in address 24) is also
available where CFIR generates two I outputs rather than
one I and one Q. See Section 3.4 for details.
to:
PFIR_GAIN = PFIR_SUM
---------------------------65536
, where PFIR_SUM is the sum of the
63 coefficients.
The 63 coefficients are identified as coefficients h0
through h62, where h31 is the center tap. The coefficients are
assumed to be symmetric, so only the first 32 coefficients (h0
3.3.6 The Second Decimate By 2 Filter
through h31) are loaded into the chip. A non-symmetric mode
(PFIR)
(NO_SYM_PFIR in address 26) allows the user to download
a 32 tap non-symmetric filter as taps h0 through h31. The
The second stage decimate by two filter is a 63 tap
decimate by 2 filter with programmable 16 bit coefficients.
Fine gain is applied at the output of the PFIR and rounded to
24 bits. Overflows are detected and hard limited. Overflows
can be directed to the channel overflow detection block.
newest sample is multiplied by h0 and the oldest is multiplied
by h31. NOTE: Filters normally multiply h0 by the oldest data,
hence one may wish to reverse the tap order in the
non-symmetric mode.
PFIR coefficient sets for a variety of standards in cellular
and cable modem applications are given in Section 7. The
PFIR filter passband must be flat in the region of the signal of
interest, and have the desired out of band rejection in the
region that will alias into the signal’s bandwidth after
decimation. Figure 11 below illustrates the passband and
stopband requirements of the filter. FPFIR is the input sample
The output samples can be multiplied by +/- 1 using a 4
bit control (NEG_CTL in register 24). If the LSB of NEG_CTL
is a 1 the even time real words are multiplied by -1. The
second LSB controls the even time imaginary words. The
third controls the odd time real and the fourth the odd time
imaginary. These can be used to invert the signal
(NEG_CTL=0xF), frequency shift by Fs/2 (NEG_CTL=0x3),
or flip the spectrum (NEG_CTL=0xA). It is also used in
complex to real conversion (see the following section).
rate to the PFIR filter. FPFIR/2 is the output sample rate of the
channel before resampling.
The externally downloaded coefficients can be used to
tailor the spectral response to the user’s needs. For example,
it can be programmed as a Nyquist (typically a
root-raised-cosine) filter for matched filtering of digital data.
The user downloaded filter coefficients are 16 bit 2’s
complement numbers. Unity gain will be achieved through
A fine gain is applied to the data after filtering and
negation. The fine gain is a positive 14 bit integer (FINE)
ranging from 0 to 16383. Fine gain is equal to:
FINE . Setting FINE to zero will clear the
FINE_GAIN = ---------------
1024
channel. The GAIN_SYNC control in address 22 can be used
Signal of Interest Passband (Typically 0.125 to 0.2 of FPFIR)
0 dB
Narrow transition band
Power
Stopband (Typically starts between 0.3 and 0.375 FPFIR)
-100 dB
0
FPFIR/4
FPFIR/2
Frequency
Figure 11. Typical PFIR Specifications
 GRAYCHIP,INC.
-9-
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
to determine when a new fine gain setting takes effect. It is
normally set to take effect immediately, but can be used to
synchronize the gain changes between multiple channels in
a beam forming system.
PFIR also contains a one PFIR sample input delay that
is independently programmable for the I and Q paths and for
the four channels (QDLY_PFIR and IDLY_PFIR in address
26). The delay is used when combining channels to allow
wider output bandwidth and/or more oversampling. It is also
used for complex to real conversion in PFIR.
3.3.7 Complex to Real Mode
PFIR can be used to convert from the complex format to
a real format for users desiring real output. This mode uses
PFIR to low pass filter the signal so it resides from -FPFIR/4 to
FPFIR /4, where FPFIR is the sample rate into the PFIR (in this
mode PFIR does not decimate by two so it is also the output
sample rate). The signal is then mixed up by FPFIR/4 (so it
now resides from 0 to FPFIR/2) and the imaginary portion is
discarded. In implementation, the samples to be discarded
are never generated. This is implemented by delaying the I
data by 1 input sample (set IDLY_PFIR in address 26 of the
channel’s control page) and mixing the PFIR output with
FPFIR /4 (set NEG_CTL=6 in address 24 of the channel’s
control page). The QONLY flag in address 24 of the
channel’s control page also needs set (The Q-half, rather
than the I-half is used due to hardware details). When
outputting real data the sample rate out of the PFIR filter is
FCK/(2N).
The REAL_ONLY bit in address 18 of the output control
page can be set to tell the output section to output real
samples. If this bit is not set, then the output will be in a
complex format with the real samples in the Q-part of the
complex word. The REAL_ONLY mode reduces the data
rate out of the chip. Note that the REAL_ONLY bit effects all
channels, so the chip does not support some channels
outputting complex data while other channels output real
data.
3.4
MULTICHANNEL MODES
The GC4016 chip contains four independent channels
and each channel can output complex sample rates up to
Fck/321. Depending on the filter coefficients and the clock
rate, the maximum single channel output bandwidth is just
under 2MHz. This output bandwidth can be doubled by
1. The Resampler (Section 3.5) can be used to increase this rate for
oversampling.
 GRAYCHIP,INC.
combining two channels using the SPLITIQ mode. Four
channels may be combined to provide 3 to 4 times the single
channel output bandwidth (with reduced out of band
rejection). The two or four multichannel modes may also be
used in the complex to real mode described in Section 3.4.6.
Two channels can also be combined to process complex
input data, thereby doubling the input bandwidth going into
the chip. 2 Four channels can be combined to both process
complex input data and to double the output bandwidth of the
chip.
Combined channels are assigned to a single resampler
or output channel using the channel map controls
(CHAN_MAP_A, B, C and D) in the resampler’s control page.
Finally, for digital modulation formats, the resampler
allows the downconverter’s output sample rate to be
between 1.5 and 2 samples per baud (symbol). The
resampler will then up-sample this to exactly 2 or 4 samples
per baud. This allows the GC4016 chip to downconvert one
4X oversampled signal at symbol rates up to 8 MBaud, or two
4X oversampled signals at 3MBaud, or four 4X oversampled
signals at 1.5MBaud (assuming a 100MHz clock rate).
3.4.1 Double Bandwidth Downconverter
Mode (SplitI/Q Mode)
Two channels work together in the SplitI/Q mode to
double the output bandwidth of the downconverter. In the
splitI/Q mode the real half of the complex output data is
processed in one channel and the imaginary half in the other.
In the splitIQ mode the CIC has a minimum decimation of 4
instead of 8, which allows channel output sample rates up to
FCK/16. The two channels being combined in the splitI/Q
mode should be programmed identically, including the tuning
frequency, except that the imaginary channel should have a
+90 degree phase shift (PHASE=0x4000). The IONLY bit in
the real channel should be set and the QONLY bit in the
imaginary channel should be set. This mode is used in the
example UMTS configuration described in Section 7.12.
Typically the chip is configured in the SPLITIQ mode so
that channels A and B are combined as one downconverter
and C and D are combined as the other. Mixed modes may
be used, such as having A and B used as narrowband
downconverters while C and D are combined into a double
bandwidth
downconverter.
The
resampler’s
CHAN_MAP_A,B,C and D controls must be set so that the
combined channels point to the same resampler channel.
2. The GC2011A digital filter chip can be used to convert sample rates up to
200MSPS into complex sample rates up to 100 MSPS. One GC2011A chip will
convert 12 bit data, two chips will convert 24 bit data.
- 10 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Double rate real output can be generated by combining
SPLITIQ and complex to real conversion. The complex to
real signal processing is the same as described in Section
3.3.7. Here however, one channel contains the real portion of
the signal and another contains the complex portion. To
convert this complex data to real, the real channel must be
delayed (set QDLY_PFIR=1), multiplied by a pattern 1,-1, 1,
-1 (set NEG_CTL=5), and QONLY should be set. The
imaginary channel should be multiplied by the sequence
-1,1,-1,1 (set NEG_CTL=10) and QONLY should be set. The
output is then available as the Q data in the selected
resampler and output channel (see CHAN_MAP_A,B,C and
D in the resampler’s control page, also see REAL_ONLY).
3.4.2 Wideband Downconvert Mode
Even wider output bandwidth is possible by combining
all four channels. This is done by noticing that PFIR
decimates the signal by two. If one pair of channels in the
SPLITIQ mode are used to generate even time sample
outputs and the other pair are used to generate odd time
sample outputs, then the PFIR filter effectively does not
decimate the signal. This allows a wider bandwidth filter to be
used in the CFIR and PFIR. Adjacent channel rejection,
however, will be reduced due to the increase in the output
bandwidth relative to the CIC’s stopbands. Also the increase
in output bandwidth makes it harder for the CFIR and PFIR
to achieve deep stop bands. Fortunately, signals that require
wideband processing are also typically signals that do not
require as much stop band rejection as narrowband signals
such as GSM.
The wide output mode uses the chip in the SPLITIQ
mode described in Section 3.5.1. In addition the QDLY_PFIR
bits are set in channels A and B. The delay of one input
sample into the PFIR will offset the decimate by two
operation so that the channel A and B outputs are the real
and imaginary parts of the even time samples and the C and
D outputs are the real and imaginary parts of the odd time
samples. This mode is used in the example UMTS
configuration described in Section 7.12.
Performing real to complex conversion in this mode
involves setting the controls as above, but also setting
QDLY_CFIR on channels A and C; setting NEG_CTL to 15
for channels B and C; setting QONLY=1 (and IONLY=0) for
all channels.
one channel while the imaginary portion (QIN) is processed
in the next channel. Channels A and B are described here,
channels C and D can be combined as well. The desired
and
mixer
output
is
IOUT=(IIN*cos-QIN*sin)
QOUT=(IIN*sin+QIN*cos). Channel A is used in the normal
mode and will output (IIN*cos) and (IIN*sin). A 90 degree
offset (PHASE=0x4000) in channel B will cause channel B to
output (-QIN*sin) and (QIN*cos). The desired result is
achieved by adding the channel A outputs to the channel B
outputs in the resampler. The resampler is told to add the
channels together by setting the ADD_A_TO_B bit for
channel A in the resampler’s control page.
The complex to real mode described in Section 3.3.7
can be used in conjunction with the complex input mode. If
both channels are configured as described in Section 3.3.7,
then the combined complex input downconverter will be in
the complex to real mode.
3.4.4 Complex Input, Double Bandwidth
(SplitI/Q) Mode
The complex input and SPLITIQ modes can be
combined. Channels A and D will process the IIN inputs.
Channels B and C will process the QIN input data. Channel A
will output (IIN*cos). Channel B will output (-QIN*sin) if its
phase is set to 90 degrees (PHASE=0x4000). Channel C will
output (QIN*cos). Channel D will output (IIN*sin) if its phase is
set to -90 degrees (PHASE=0xC000). The I OUT output is
formed by adding channel A to channel B, and the QOUT
output is formed by adding channel C to channel D. The
IONLY bits must be set for channels A and B, and the
QONLY bits must be set for channels C and D.
The complex to real mode can be added to this mode by
delaying the I word by half (set QDLY_PFIR in channels A
and B) and mixing by Fs/4 (set NEG_CTL to 5 in channels A
& B and set NEG_CTL to 10 in channels C & D). QONLY
instead of IONLY needs to be set in channels A and B.
3.4.5 Multichannel Summary
Table 2 summarizes the settings for the eight
multichannel modes. A phase setting of 90 indicates a +90
degree phase offset, or a value of 0x4000. A phase setting of
-90 indicates a -90 degree phase offset, or a value of
0xC000.
3.4.3 Complex Input, Narrowband output
Complex input data can be processed using two
channels. The real portion of the input (IIN) is processed in
 GRAYCHIP,INC.
- 11 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Table 2: Multichannel mode settings
Input
Real
Output
Rate
Two 2X output channels
Single 4x output channel
1x
Output
Format
Complex
Complex
Complex
Channel
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
A
B
C
D
SPLITIQ
PHASE
QDLY_CFIR
IDLY_PFIR
QDLY_PFIR
NEG_CTL
IONLY
QONLY
ADD_TO
CHAN_MAP
1
0
0
0
0
0
1
0
0
0
1
90
0
0
0
0
0
1
0
0
1
0
0
0
0
0
1
0
0
1
1
90
0
0
0
0
0
1
0
1
1
0
0
0
1
5
0
1
0
0
1
90
0
0
0
10
0
1
0
0
1
0
0
0
1
5
0
1
0
1
1
90
0
0
0
10
0
1
0
1
1
0
0
0
1
0
1
0
0
0
1
90
0
0
1
0
0
1
0
0
1
0
0
0
0
0
1
0
0
0
1
90
0
0
0
0
0
1
0
0
1
0
1
0
1
0
0
1
0
0
1
90
0
0
1
15
0
1
0
0
1
0
1
0
0
15
0
1
0
0
1
90
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0
90
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
90
0
0
0
0
0
0
0
1
0
0
0
1
0
6
0
1
1
0
0
90
0
1
0
6
0
1
0
0
0
0
0
1
0
6
0
1
1
1
0
90
0
1
0
6
0
1
0
1
1
0
0
0
0
0
1
0
1
0
1
90
0
0
0
0
1
0
0
0
1
0
0
0
0
0
0
1
1
0
1
-90
0
0
0
0
0
1
0
0
1
0
0
0
1
5
0
1
1
0
1
90
0
0
1
5
0
1
0
0
1
0
0
0
0
10
0
1
1
0
1
-90
0
0
0
10
0
1
0
0
3.5
Complex
Real
Real
RESAMPLER
around 1.5 to 2 times the signal’s bandwidth1. The resampler
is then used to increase the sample rate up to the required
2X, 4X or 8X rate.
Section 7.7 shows example resampler configurations
and their performance.
The resampler can also be used as an additional filter to
optimize the passband or stopband response of the channel.
1. The PFIR’s 63 tap filter’s stopband, transition band and passband ripple
performance improves as its output rate decreases relative to the signal’s
bandwidth. The resampler’s performance, however, begins to decrease when
the PFIR output rate is below 1.5 times the signal’s bandwidth.
Complex
Real
The resampler consists of an input buffer, an
interpolation filter, and a final shift block. A functional block
diagram of the resampler is shown in Figure 12.
The resampler’s sampling rate change is the ratio
NDELAY/NDEC where NDELAY and NDEC are the
interpolation and decimation factors shown in Figure 12. The
decimation amount NDEC is a mixed integer/fractional
number. When NDEC is an integer, then the exact sampling
instance is computed and there is no phase jitter. If NDEC is
fractional, then the desired sampling instance will not be one
of the possible NDELAY interpolated values. Instead the
nearest interpolated sample is used. This introduces a timing
error (jitter) of no more than 1/(2*NDELAY) times the input
sample period.
The input buffer accepts 24 bit data from the four input
channels, and adds them as necessary to form 1,2, or 4
resampler channels (see the ADD_TO control bits in address
21 of the resampler control page). The input buffer serves
both as a FIFO between the channels and the resampler,
and as a data delay line for the interpolation filter. The 64
complex word input buffer can be configured as four
segments of 16 complex words each to support 4 resampler
Interpolation Filter
INPUT
BUFFER
Real
3.5.1 Functional Description
The resampler will independently filter and change the
data rate of each channel. The most common application of
the resampler is to increase the sample rate of the data so
that it will match a desired symbol or bit rate. Demodulators
for digital modulation schemes, such as GMSK, QPSK, QAM
or CDMA, for example, require sample rates which are 1X,
2X, 4X or 8X times the bit or symbol rate of the modulation.
In these cases, the maximum down converter filter
performance is achieved when the PFIR output rate is
IN
2x
Fractional
Decimation
NDEC
NDELAY
QTAP Filter Coefs
FINAL
SHIFT
AND
ROUND
OUT
Resampling Filter
Figure 12. Resampler Channel Block Diagram
 GRAYCHIP,INC.
- 12 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
the zero padded data in order to suppress the interpolation
images. The last spectrum shows the final result after
channels, or as two segments of 32 complex words each to
support 2 resampler channels, or as a single segment of 64
complex words to support a single resampler channel. The
number of segments is set by NCHAN in address 16 of the
resampler control page.
decimating by NDEC.
3.5.2 The Resampler Filter
The interpolation filter zero pads the input data by a
factor of NDELAY and then filters the zero padded data using
a QTAP length filter. The output of the QTAP filter is then
decimated by a factor of NDEC.
Figure 13(b) illustrates the spectral shape requirements
of the QTAP filter. If the desired signal bandwidth is B, then
the filter’s passband must be flat out to B/2 and the filter’s
stop band must start before FS-B/2. The user designs this
The resampling ratio for each channel is determined by
setting the 32 bit RATIO controls in addresses 16 through 31
of the resampler ratio page. The value of RATIO is defined
as:
RATIO = 2
filter assuming a sample rate equal to NDELAY*FS. Section
7.7 contains example resampler filters coefficient sets. Other
passband and stopband responses can be used, such as
root raised cosine receive filters, as desired. The resampler
filter can also be used to augment the CIC, CFIR and PFIR
filters’ spectral response.
26  NDEC 
26 INPUT SAMPLE RATE
------------------------ = 2  --------------------------------------------------------------------
 NDELAY
 OUTPUT SAMPLE RATE
The number of filter coefficients, QTAP, is equal to
NMULT*NDELAY, where NMULT is the number of multiplies
available to compute each resampler output, and NDELAY is
either 4, 8, 16, 32 or 64 as will be described later. The
maximum filter length is 512. The user specifies NMULT in
address 17 of the resampler control page.
Up to four ratios can be stored within the chip. A ratio
map register (address 23) selects which ratio is used by each
channel.
The three spectral plots shown in Figure 13 illustrate the
steps required to resample the channel data. The first
spectral plot shows the data just after zero padding. The
sample rate after zero padding is NDELAY*FS, where FS is
The filter can be symmetric, or non-symmetric, as
selected by the NO_SYM_RES control in address 17 of the
resampler control page. The symmetric filter is of even length
which means the center tap repeats.
the sample rate into the resampler. The second spectrum
shows the shape of the QTAP filter which must be applied to
NDELAY Images added by zero padding
Power
...
...
Freq
-3FS
-2FS
-FS
-FS/2
FS/2
FS
2FS
3FS
a) Spectrum after zero padding
QTAP Filter
Frequency Response
Power
Images reduced
by Filtering
...
...
-3FS
-2FS
-FS
-FS/2
FS/2
FS
2FS
Freq
3FS
b) Spectrum after filtering
In-band aliasing caused by decimation by NDEC
Power
-FO
-FO /2
FO/2
Freq
FO =(NDELAY/NDEC)FS
c) Final Spectrum after decimating by NDEC
Figure 13. The Resampler’s Spectral Response
 GRAYCHIP,INC.
- 13 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
The 12 bit filter coefficients are stored in a 256 word
memory which can be divided into one, two, or four equal
blocks. This allows the user to store one symmetric filter of
up to 512 taps, two symmetric filters of up to 256 taps each,
or four symmetric filters of up to 128 taps each. The number
of filters is set by NFILTER in address 16 of the resampler
control page. The filter used by each channel is selected
using the FILTER_SEL controls in address 18 of the
resampler control page. The filter lengths are cut in half if the
filters are not symmetric. The coefficients are stored in
memory with h0 stored in the lowest address, where h0 is the
coefficient multiplied by the newest piece of data. The center
tap of a symmetric filter is h (QTAP/2)-1. The coefficients for
multiple filters (NFILTER>1), are interleaved in the 256 word
memory.
3.5.3 Restrictions on NMULT
The user does not directly set the value of NDELAY. The
chip sets the value of NDELAY using NO_SYM_RES,
NMULT and NFILTER according to:
NDELAY
( 2 – NO_SYM_RES )
= Floor_2 [ 256 ---------------------------------------------]
( NMULT ) ( NFILTER )
where the function FLOOR_2[X] means the power of two
value that is equal to or less than “X”. Since NMULT is
restricted to be greater than or equal to 6 and less than or
equal to 64, then NDELAY is either 4, 8, 16, 32 or 64. The
length of the filter is then:
The value of NMULT determines both the length of the
filter and the number of delays in the resampling operation.
In general one would choose the largest value of NMULT
which gives an adequately large value of NDELAY. The
choice of NMULT, however, must meet several restrictions.
NMULT must be greater than a minimum, it cannot exceed
the available number of multiplier cycles, and it must be less
than the input delay line segment size. These restrictions are
described below.
The minimum value of NMULT is determined by the
minimum number of clock cycles required to update the
resampler’s state. This is a hardware restriction imposed by
the chip’s architecture. This limitation is:
if there are two or more outputs
NMULT ≥ 7
if there is only one output
NMULT is the number of complex multiplier operations
required to compute an output sample. Since the resampler
can perform two multiplies every clock cycle, the value of
NMULT cannot exceed two times the number of clock cycles
 GRAYCHIP,INC.
∑
Note that the resampler’s output sample rate is usually
much less than the clock rate, so that NMULT is rarely limited
by this restriction. NOTE: THE NUMBER OF CLOCK
CYCLES AVAILABLE TO THE RESAMPLER IS REDUCED
BY THE CLOCK DIVIDER DISCUSSED IN THE NEXT
SECTION.
The value of NMULT must also be less than the size of
the delay line formed by the input buffer. The size of the delay
line is either 16 for four resampler channels, 32 for two
channels or 64 for a single channel as set by the NCHAN
control in address 16 of the resampler control page. This
limits NMULT to be less than or equal to 15, 31 or 63
dependent upon the number of resampler channels1.
The typical resampler configuration will have four active
channels, all using the same filter and the same resampling
ratios. The typical configuration has NCHAN set to 4,
NFILTER set to 1, NMULT set to 15 and NO_SYM_RES set
to 0. This sets NDELAY to 32 and QTAPS to 480. See
Section 7.7.
3.5.4 The Resampler Clock Divider
QTAP = ( NDELAY ) ( NMULT )
NMULT ≥ 6
available to the resampler for each channel. The number of
clock cycles available to the resampler is equal to the clock
rate to the chip divided by the sum of the output sample rates
for each resampler channel. Hence, NMULT must satisfy:
F CK
NMULT ≤ 2 × -----------------------------------------------------------( OUTPUT RATES )
The resampler has a clock divider that can be used to
reduce power consumption and to slow the calculation rate
of the resampler. The clock divider divides the internal clock
by factors of one to 256 using the RES_CLK_DIV control in
address 22 of the resampler control page. The clock divider
reduces the resampler’s computational throughput, so care
must be taken not to reduce the clock rate to the point that
the computations can not be completed on time.
3.5.5 Final Shift and Round
The gain of each resampler output is adjusted by an
up-shift by 0-15 bits (FINAL_SHIFT). This up-shift is applied
just before rounding to the 12, 16, 20 or 24 MSBs (ROUND).
The values of FINAL_SHIFT and ROUND are set in control
register 19 of the resampler control page. The resampler
1. NOTE: If the resampler is being used at much less than its maximum
capacity, i.e., NMULT is much less than twice the number of clock cycles
available (See also RES_CLK_DIV), AND the channels are synchronous, then
NMULT may equal the size of the delay line.
- 14 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
3.7
gain is:
RES_SUM
RES_GAIN = ( ----------------------------------------------)(2
32768 × NDELAY
FINAL_SHIFT
)
where RES_SUM is the sum of the QTAP coefficients.
3.5.6 By-Passing the Resampler
The resampler is bypassed by using a configuration
which has h0 set to 1024, all other taps set to zero, NMULT
set to 7, NO_SYM_RES set to 1, FINAL_SHIFT set to 5, and
RATIO set to 226 (0x04000000). Note that the NDELAY term
in the RES_GAIN equation shown above does not apply in
this case and should be set to unity in the gain equation.
3.5.7 Adaptive Ratio Change
The ratio maps (address 23) can be used to dynamically
adjust each channel’s resampling ratio. Four different ratios
can be loaded into the chip. The ratio map indicates which
ratio is to be used by which channel. This feature can be
used to lock the resampling to a desired output timing by
adaptively selecting ratios which are slightly larger or smaller
or equal to the desired ratio. If the timing needs to be
accelerated, then the larger ratio is selected, if the timing is
correct, then the exact ratio is used and if the timing needs to
be retarded, then a smaller ratio is used. The ratio change
can be synchronized using the RATIO_SYNC control in
address 21.
3.6
OVERALL GAIN
The overall gain of the chip is a function of the amount
of blanking (NZEROS), the decimation programmed into the
chip (N), the scale circuit setting in the CIC filter (SHIFT,
SCALE and BIG_SCALE), the coarse gain setting
(COARSE), the sum of the CFIR coefficients (CFIR_SUM),
the sum of the PFIR coefficients (PFIR_SUM), the fine gain
(FINE_GAIN), the sum of the resampler coefficients
(RES_SUM), the number of delays in the resampler filter
(NDELAY) and the final shift (FINAL_SHIFT). The overall
gain is shown below. The term in braces should be less than
or equal to unity. Note that the NDELAY term is unity if the
resampler is bypassed (see Section 3.5.6).
Unity gain is defined relative to the MSB of the I/O data.
Unity gain means that a maximum DC input with a tuning
frequency of zero, will produce a maximum DC output,
independent of the I/O words sizes.
PEAK COUNTER
Each channel has a peak counter which can monitor the
input signal’s strength, or can count the number of overflows
detected at three points within the channel. The peak counter
can count overflows due to the COARSE_GAIN circuit, it can
count overflows at the output from the CFIR, and it can count
overflows at the output of the PFIR.
If PEAK_MODE is set to 0 in address 28 of the channel
control page, then the peak counter will monitor the signal
strength at the input to the channel. The magnitude of each
input sample is compared against a threshold selected by the
PEAK_THRESH control in the same address. The threshold
can be 1/4, 1/2 3/4 or full scale. The peak counter increments
each time the magnitude equals or exceeds the threshold.
The input threshold crossings can be used in an AGC loop to
optimize the input signal level to the ADC.
If PEAK_MODE is 1, then the peak counter monitors
overflows at points selected by the PEAK_SELECT control in
address 26 of the channel control page. The peak counter
increments each time an overflow is detected. Note that the
data value is saturated to plus or minus full scale, as
appropriate, when the overflow occurs. The overflow count
can be used to check if the signal gain is set too high
anywhere along the processing chain.
The counter counts up to 255 and stops. The counter
value is saved in a holding register and the counter is cleared
by the sync selected by the PEAK_SYNC signal (see register
28). The sync can either be one of the chip’s input syncs, a
one-shot pulse, or the terminal count of the chip’s sync
counter. A periodic sync allows the user to count threshold
crossings or overflows during fixed time periods.
The user reads the count through the PEAK_COUNT
control register at address 29 of the channel control page.
The holding register contents are transferred to the
PEAK_COUNT register a clock cycle after the user reads the
PEAK_COUNT register. This transfer is delayed to prevent
the user from reading a peak count value at the same time
the circuit is updating the register.
Note that there are two other points in the chip where
overflows can occur. The first is at the input to the resampler
where samples from two channels can be added together as
when processing complex input data. This may cause an
overflow even though the samples feeding the resampler did
not overflow. If an overflow occurs here, the sample is
saturated and a control bit in the status register (address 1)

5 ( SHIFT + SCALE + 6 × BIG_SCALE – 62 )  COARSE CFIR_SUM PFIR_SUM FINE_GAIN
FINAL_SHIFT
1
RES_SUM )N 2
) ( ------------------) ( ------------------ ) ( -------------------) ( --------------------------------------------)(2
)
GAIN =  ( --------------------( 2
32768 × NDELAY
NZEROS + 1
65536
65536
1024


 GRAYCHIP,INC.
- 15 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
is set. The user can clear this bit by writing a 0 to the register.
The control bit indicates that at least one sample overflowed
since the last time the bit was cleared. Since the datapath
going into the resampler is 24 bits, it is recommended that the
gain be set such that this overflow never occurs. The second
overflow point is at the resampler output. The user can check
for these overflows by examining the chip’s output data for
maximum (saturated) values.
3.8
OUTPUT MODES
Data from the resampler can be output in one of several
modes:
(1) The microprocessor mode where the outputs
are read from the 8 bit control port,
The chip supports different decimation ratios and/or
different resampling ratios for each channel. If different ratios
are used, or if the channels have not been synchronized
using the DEC_SYNC controls (See Section 3.11), then the
channels are asynchronous and tag bits are required in order
to sort out the channel data. If the same ratios are used, and
the channels have been synchronized, then the output is
synchronous and the tag bits are not required.
If the complex to real conversion modes of the PFIR
described in Section 3.3.7 are used, then the REAL_ONLY
control bit in address 18 can be used to output real rather
than complex data. The REAL_ONLY control can only be
used if all of the channels are in the complex to real mode, a
mix of complex and real data is not allowed. If the complex
output format is used to output real data (REAL_ONLY=0),
then the real outputs are output as the Q-half of the complex
output words. The I-half should be ignored.
(2) The wide word microprocessor mode where
the outputs are read as 32 bit integers from a
wide word control port,
(3) The synchronous serial mode where the
outputs are output through one or more of the
four serial ports,
(4) The asynchronous serial mode where the
channels are output on the same serial stream,
but have their own serial frame sync,
The suggested control register settings for each of the
output modes are shown in Table 3. These settings assume
all four channels are active and the outputs are synchronous.
(5) The nibble mode where the outputs are output
in a 4-bit at a time nibble serial format,
(6) The link mode which is compatible with the four
bit SHARC link port, or
(7) The parallel mode which outputs the data as 24
bit words.
Only one mode is supported at a time. The output
modes are controlled by writing to addresses 16 through 26
within page 98 (the output control page).
A FIFO memory holds blocks of resampler output data,
where each block contains one, two or four complex samples
as specified by the BLOCK_SIZE control in address 20. The
FIFO is used to buffer the data rate between the resampler
and the output. The number of blocks held in the FIFO
depends upon the output mode being used.
The data from the resampler can be rounded to 12,16,
20 or 24 bits. The BITS_PER_WORD control in address 20
will set the serial mode word size to be between 12 and 32
bits, or to set nibble and link mode words to be either 16 or
32 bits. If the output word size is larger than the resampler’s
output size, then the unused LSBs are cleared.
Tag bits (four bits per tag) may be programmed to
replace the four LSBs of the specified output word size. The
tags are used to identify the source of the data. The user can
program eight different tag values in addresses 23, 24, 25
 GRAYCHIP,INC.
and 26, two for each of the four complex output words. Tags
are enabled using the TAG_EN control bit in address 17. A
special 2 bit tag mode control (TAG_22 in address 19 of the
resampler control page) can be used to output 24 bit words
containing 22 bits of data and 2 bits of tag.
The output circuitry is reset upon power up. It is enabled
by clearing the OUT_BLK_RESET and PAD_RESET bits in
address 0. See Section 3.12 for the proper initialization
procedures.
An output block sync is provided (OUT_BLK_SYNC in
address 17) which can be used to synchronize the serial
clocks of multiple chips. See Section 3.12 for the proper
synchronization sequences.
3.8.1 Microprocessor Mode
In the microprocessor mode (OUTPUT_MODE=0 in
address 18) the outputs are read in bytes from the control
port. The outputs are accessed by reading addresses 16
through 31 in pages 96 and 97.
The output FIFO buffers two blocks of complex samples
in this mode. One block is accessed through the
microprocessor port as another block is being filled with new
data. When the new block is filled, the FIFO swaps the blocks
so that the user can access the new data. The block size is
usually set to buffer four complex samples per block
(BLOCK_SIZE=3). If only one or two channels are being
output, then the block size may be reduced, but it is
suggested to keep it at four samples per block in order to
- 16 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Table 3: Output Mode Controls
Suggested Default for Each Mode (hex values)
Control Register
Address
(Page 98)
Microprocessor
mode
Wide Word
microprocessor
mode
Serial modes
Four serial
outputs
Nibble mode
Link mode
Parallel
mode
FF
(synchronous)
Asynchronous Mode
Tristate Controls
16
02
FA
7F
7F1
7F1
7F1
7D
Output Formats
17
40
40
40
40
40
40
48
40
Output Modes
18
08
08
28
282
282
4A
CB
6C
TDM output
Output Frame Controls
19
00
00
01
073
01
07
00
00
Output Word Sizes
20
E8
E8
E9
EF
29
C8
08
E85
Output Clocks
21
00
00
B1
01
01
01
01
01
Serial Mux Controls
22
00
00
E4
E4
E4
00
00
00
Output Tag A
23
10
00
10
Output Tag B
24
32
11
32
Output Tag C
25
45
22
45
Output Tag D
26
67
33
67
Miscellaneous
28
02
034
02
NOTES: 1. For TDM serial or nibble modes the master chip’s value should be 7F, the slaves’ 7D.
2. For TDM serial or nibble modes the master chip’s value should be 28, the slaves’ 20.
3. For multi-chip modes the frame length is set as described in Section 3.8.3.
4. Also set the RND22control bit in address 19 (bit 6) of the resampler control page (page 64).
5. For Asynchronous parallel modes set address 20 to 28.
reduce the interrupt rate and to let the user read more
samples after each interrupt.
When the user has finished reading the block of data he
sets the READY control bit in the Status Register (address1).
When the output FIFO has a new block of data ready, it will
clear the READY bit. If a block of data completes and the
READY bit is not set, then the chip sets the MISSED control
bit. The MISSED bit serves as a diagnostic indicating that the
processor was too slow and missed a block of data. Note that
the rate at which the blocks are filled, and therefore the rate
at which the data needs to be read from the microprocessor
port, is dependent upon how fast the resampler can generate
new output samples. When the resampler is interpolating,
which is the resampler’s most commonly used mode, it will
generate multiple output samples each time it receives a
sample from a downconverter channel. The user should use
the resampler’s clock divider to slow its computation rate if
the resampler is generating outputs too fast (See Section
3.5.4).
The RDY pin can serve as an interrupt to inform the
processor that a new block of samples is ready. The width
and polarity of the RDY pulse is programmable (See
EN_RDY in address16 and INV_RDY and RDY_WIDTH in
address 17).
The mapping of the output data into registers 16 through
31 is dependent upon the OUTPUT_ORDER control mode in
address 21, the number of active channels, and whether the
channels are synchronous or asynchronous.
If OUTPUT_ORDER=0 and BLOCK_SIZE=3, then each
block will contain the next four samples output from the
 GRAYCHIP,INC.
resampler. If the channels are synchronous, then the block
will contain one sample from each channel. If there are only
two active channels, then the block will contain two samples
from each channel. If there is only one active output channel,
then the block will contain four consecutive samples from the
active channel. If the channels are asynchronous, or if there
are three active channels, then the output block will contain
the next four samples computed by the resampler. In this
case tags are required to separate the channels.
OUTPUT_ORDER=0 must be used for asynchronous data.
If OUTPUT_ORDER=1 or 2, then the channels are
written into the FIFO using the channel number as part of the
address. In this mode the previous block is complete when
the Q-half of the output from channel A arrives. If the
resampler ratio is the same integral value for all channels,
then channels B, C and D can be powered down and back up
again while maintaining proper channel ordering. Channel A
may not be powered down because it provides the “Block
Complete” timing signal. OUTPUT_ORDER settings of 1 and
2 are valid for synchronous channels, but NOT for
asynchronous channels.
The following tables illustrate how the 24 bit channel
outputs are mapped into addresses 16 through 31 of pages
96 and 97. Note that addresses 16, 20, 24 and 28 are the
unused LSBs of the 32 bit words and always read back zero.
Addresses 19, 23, 27 and 31 contain the MSBs of the output.
In the 12,16, or 20 bit output mode, all values are rounded
into the MSBs and the unused LSBs are cleared. These
tables assume that BLOCK_SIZE = 3 which sets the block
size to four complex words.
- 17 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Table 4: Channel Output Map, Synchronous Four Channel Mode
CHANNEL OUTPUT PAGE 96
ADDRESSES
OUTPUT_ORDER=0
OUTPUT_ORDER=1
OUTPUT_ORDER=2
16,17,18,19
AOUT, I-half
AOUT, I-half
AOUT, I-half
20,21,22,23
AOUT, Q-half
BOUT, I-half
BOUT, I-half
24,25,26,27
BOUT, I-half
AOUT, Q-half
COUT, I-half
28,29,30,31
BOUT, Q-half
BOUT, Q-half
DOUT, I-half
CHANNEL OUTPUT PAGE 97
ADDRESSES
OUTPUT_ORDER=0
OUTPUT_ORDER=1
OUTPUT_ORDER=2
16,17,18,19
COUT, I-half
COUT, I-half
AOUT, Q-half
20,21,22,23
COUT, Q-half
DOUT, I-half
BOUT, Q-half
24,25,26,27
DOUT, I-half
COUT, Q-half
COUT, Q-half
28,29,30,31
DOUT, Q-half
DOUT, Q-half
DOUT, Q-half
Table 5: Channel Output Map, Synchronous Two Channel Mode
CHANNEL OUTPUT PAGE 96
ADDRESSES
OUTPUT_ORDER=0
OUTPUT_ORDER=1
OUTPUT_ORDER=2
16,17,18,19
A(i)OUT, I-half
AOUT, I-half
AOUT, I-half
20,21,22,23
A(i)OUT, Q-half
BOUT, I-half
BOUT, I-half
24,25,26,27
B(i)OUT, I-half
AOUT, Q-half
unused
28,29,30,31
B(i)OUT, Q-half
BOUT, Q-half
CHANNEL OUTPUT PAGE 97
ADDRESSES
OUTPUT_ORDER=0
OUTPUT_ORDER=1
OUTPUT_ORDER=2
16,17,18,19
A(i+1)OUT, I-half
unused
AOUT, Q-half
20,21,22,23
A(i+1)OUT, Q-half
BOUT, Q-half
24,25,26,27
B(i+1)OUT, I-half
unused
28,29,30,31
B(i+1)OUT, Q-half
Table 6: Channel Output Map, Single Channel Mode
CHANNEL OUTPUT PAGE 96
ADDRESSES
OUTPUT_ORDER=0
OUTPUT_ORDER=1
OUTPUT_ORDER=2
16,17,18,19
A(i)OUT, I-half
AOUT, I-half
AOUT, I-half
20,21,22,23
A(i)OUT, Q-half
unused
unused
24,25,26,27
A(i+1)OUT, I-half
AOUT, Q-half
28,29,30,31
A(i+1)OUT, Q-half
unused
CHANNEL OUTPUT PAGE 97
ADDRESSES
OUTPUT_ORDER=0
OUTPUT_ORDER=1
OUTPUT_ORDER=2
16,17,18,19
A(i+2)OUT, I-half
unused
AOUT, Q-half
20,21,22,23
A(i+2)OUT, Q-half
24,25,26,27
A(i+3)OUT, I-half
28,29,30,31
A(i+3)OUT, Q-half
 GRAYCHIP,INC.
unused
- 18 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
In the complex to real conversion modes (see Section
3.3.7), the real output is available in the Q portion of the
registers and the I portion is invalid. The REAL_ONLY
control has no effect upon the microprocessor mode.
Tables 4, 5 and 6 assume the block size control
(BLOCK_SIZE in address 20) is set to four samples. If the
block size is two and OUTPUT_ORDER is 0, then page 97 is
unused. If the block size is one and OUTPUT_ORDER is 0,
then only addresses 16 through 23 of page 96 are used.
3.8.2 Wide Word Microprocessor Mode
The wide word microprocessor mode is selected by
setting OUTPUT_MODE=0 and by enabling the parallel
output port (EN_P0=EN_P1=EN_P2=EN_P3=EN_PAR=1 in
address 16). The wide word microprocessor mode allows the
user to read 32 bit words, rather than bytes. The 8 bit
microprocessor port is augmented with the 24 bit parallel port
to provide the 32 bit interface. The bit mapping into the 32 bit
port is shown below.
32 BIT Wide Word PORT
MSB
LSB
P23
P0 C7
PARALLEL PORT
(contains the 24 bit output data)
C0
CONTROL PORT
(reads back zero)
Figure 14. Wide Word Microprocessor Port
The channel outputs are read using addresses 16, 20,
24, and 28 of pages 96 and 97. The channel ordering is the
same as described in Section 3.8.1 above and shown in
Tables 4, 5 and 6.
The upper 24 bits of the wide word microprocessor port
are read only bits. The port and has the same control and
timing characteristics as shown in Figures 2 and 3 for the
normal microprocessor port.
3.8.3 Synchronous Serial Outputs
The
serial
mode
is
selected
by
setting
OUTPUT_MODE=1, MASTER=1 and enabling the serial
output pins (EN_SCK=EN_SFS=EN_P0,1,2,3=1 in address
16). The chip provides a bit serial clock (SCK), a frame strobe
(SFS) and four data bit lines (SOUT A,B,C and D) to output
the data in a serial format. The I and Q parts of complex
outputs are always multiplexed onto the same bit-serial pin.
The I-part is output first, MSB to LSB, followed by the Q-part.
In the real mode (REAL_ONLY=1), only the Q word per
channel is output. The REAL_ONLY mode is used with the
complex to real conversion mode of the PFIR.
The serial clock is the input clock (CK) divided by 1-16
(SCK_RATE control in address 21). For even divisions, the
serial clock changes on the rising edge of CK. The serial
clock has a 50% duty cycle for all divisions. The polarity of
the serial clock is programmable (which allows output data to
be sampled on the rising or falling edge of SCK).
The serial output word size is programmable to be
12,16,20,24,28, or 32 bits per word (BITS_PER_WORD
control in address 20). Note that the maximum word size out
of the resampler is 24 bits so for output word sizes of 28 or
32 bits the lower bits are always zero. For complex data
outputs, pairs of words, each being 12 to 32 bits, are output.
Synchronous channel outputs can be transmitted as one
channel per serial output on four separate bit-serial output
pins (SOUTA, SOUTB, SOUTC, and SOUTD), or
multiplexed as two channels per pin onto two output pins
(SOUTA and SOUTB), or multiplexed as all four channels on
the same pin (SOUTA) as specified by the
OUTPUT_ORDER and NSERIAL control bits in address 21.
The four channel, each on their own serial pin, mode uses
NSERIAL=3 and OUTPUT_ORDER=2. The two channels
per pin mode uses NSERIAL=1 and OUTPUT_ORDER=1.
The four channels on a single pin mode uses NSERIAL=0
and OUTPUT_ORDER=0.
The serial streams SOUTA, SOUTB, SOUTC and
SOUTD are normally output on pins P0, P1, P2 and P3,
respectfully. If required, the SMUX_0, SMUX_1, SMUX_2
and SMUX_3 controls in address 22 can be used to select
which stream is output on which of these pins.
For synchronous channels the FIFO block size
(BLOCK_SIZE) should be set to match the number of active
channels. If four channels are active, then BLOCK_SIZE
should be set to 3. If two channels are active, then
BLOCK_SIZE should be set to 1. If only one channel is
active, then BLOCK_SIZE should be set to 0.
The outputs are output in frames. Output frames start
when the previous frame has completed AND a new data
block is ready in the FIFO (See BLOCK_SIZE above). The
minimum output frame length can be programmed to be 1 to
64 words (up to 32 complex samples) using the
FRAME_LENGTH control in address 19. Longer frame sizes
can be used to time division multiplex (TDM) channels from
multiple chips onto a signal serial bus or to smooth data flow
when resampling.
The number of words output on each serial pin during a
frame can be programmed to be 1 to 8 words (1 to 4 complex
 GRAYCHIP,INC.
- 19 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
samples) using the WORDS_PER_FRAME control in
address 20. The WORDS_PER_FRAME control is usually
set to match the number of active output channels
multiplexed onto each serial pin using the NSERIAL control.
In real output modes each sample is one word. In complex
output modes each sample is two words. NOTE:
FRAME_LENGTH must be greater than or equal to
WORDS_PER_FRAME.
If the selected number of words have been output, and
the frame is not complete, or a new FIFO block is not ready,
then the frame strobe (SFS) will remain inactive, the data bits
will go tristate, and the serial clock (SCK) will continue. In this
case a new frame will start on the next SCK pulse after the
frame completes and a new FIFO block is ready. If new
outputs are ready before the previous frame is finished, then
the FIFO will buffer the new data until the previous frame is
complete. THE FRAME LENGTH AND SERIAL CLOCK
RATES MUST BE SET SO THAT THE FRAME RATE IS
GREATER THAN OR EQUAL TO THE AVERAGE
RESAMPLER OUTPUT RATE. This means that:
(Average # of clocks per resampler output) ≥
(Frame length)(bits per word)(clocks per bit)
where the average # of clocks per resampler output is equal
to:
(Average # of clocks per resampler output) =
(Channel deciation)(NDEC/NDELAY)
and NDEC/NDELAY is the resampling ratio (see Section
3.5.1).
The frame strobe signal (SFS) is programmable to come
once per frame, once per complex word, or once per word
using the SFS_MODE control in address 19. SFS is one SCK
clock cycle wide and always comes one SCK ahead of the
first output bit in the transfer. Its polarity is programmable
(INV_SFS in address 17). If the outputs are synchronous and
the frame strobe signal is once per frame, then a word’s
position in the frame relative to the SFS strobe can be used
to identify which channel each sample belongs to. If not, tags
may be used to identify channels. Some processors require
an SFS strobe with each word so the position within a frame
relative to the SFS strobe cannot be used to identify
channels. Tag bits must be used for these processors, but
may be turned off once synchronization has been achieved.
Note that frames can be generated back to back, specifically,
the frame strobe can occur at the same time as the last bit in
the previous frame.
The serial frame timing is illustrated below in Figure 15.
Synchronous data from multiple chips may be time
division multiplexed (TDM) onto the same serial bus. A
master GC4016 chip (MASTER=1 in address 18 and
 GRAYCHIP,INC.
EN_RDY=1 in address 16) provides the frame strobe signal
and serial clock and a RDY start of frame pulse. The
bidirectional RDY pin is an output pin from the master chip
and is an input pin for the slave chips (MASTER=0,
EN_RDY=0). The slave chips use the RDY input frame
strobe to identify the start of frame. The master will drive the
serial
outputs
for
the
first
1-8
words
(WORDS_PER_FRAME) of the frame, and then will tristate
its serial data out. The slave chips are programmed using the
FRAME_LENGTH control to delay the start of their outputs
by 1-63 words from the beginning of frame. Note that the
delay is programmed in words, NOT complex samples. The
delay can be programmed independently for each slave chip
so that each chip can have its own block of time in which to
output data. For example, a four chip TDM stream can be
generated, where each chip is outputting eight words (four
complex outputs), by setting WORDS_PER_FRAME=7 (8
words per chip) in all four chips, FRAME_LENGTH=31 in the
master chip (32 words per frame), FRAME_LENGTH=7 in
the second chip (start at word 8), FRAME_LENGTH=15 in
the third chip (start at word 16) and FRAME_LENGTH=23 in
the fourth chip (start at word 24).
The TDM mode requires that the serial clocks in the
master and slave chips have been synchronized using the
OUT_BLK_SYNC control. See Section 3.12 for details.
The TDM mode is intended for use with a single serial
output stream (NSERIAL=0), but will also work with one, two
or four streams. The TDM outputs are then identified within
each stream according to the NSERIAL and
OUTPUT_ORDER controls as shown in Figure 15(c).
3.8.4 Asynchronous Serial Outputs
Asynchronous channels must be output on a single
serial stream. Tag bits or separate frame strobes must be
used to identify the channels. In the asynchronous mode
each channel sample is output as a serial word (or complex
pair) as soon as the resampler has finished generating it. Tag
bits or separate frame strobes are used to match the serial
output word with the channel it came from. Asynchronous
channels must use NSERIAL=0, OUTPUT_ORDER=0, and
BLOCK_SIZE=0.
WORDS_PER_FRAME
and
FRAME_LENGTH must be set to 0 for real data and 1 for
complex data. Four bit tags are enabled by setting
TAG_EN=1 and TAG_22=0. The two bit tag mode
(TAGEN=0, TAG22=1) can be used to output 24 bit words
that are 22 bits of data plus 2 bits of tag.
Separate frame strobes are enabled by using the two bit
tag mode and setting the EN_4_FS control bit (address 28 of
- 20 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
clock is continuous
SCK
The data pins go tristate after the last bit
SFS
SOUT
I15
I14
I1
I0
Q15
Q14
Q1
Q0
(a) SFS_MODE=3: FRAME SYNC AT THE START OF EACH WORD (16 bit words shown)
SCK
SFS
I15
SOUT
I14
I1
I0
Q15
Q1
Q14
Q0
(b) SFS_MODE=2: ONE FRAME SYNC AT THE START OF EACH COMPLEX WORD
The minimum output frame length is set by FRAME_LENGTH
SFS
SOUTA
SOUTB
SOUTC
SOUTD
IA
QA
IA
IB
QB
IB
QA
QB
IC
QC
IC
QC
ID
QD
ID
QD
Four Output streams: NSERIAL=3, WORDS_PER_FRAME=1, OUTPUT_ORDER=2
SFS
SOUTA
SOUTB
SFS
SOUTA
IA
QA
QA
IB
QB
IC
QC
ID
QD
IC
QC
Two Output Streams: NSERIAL=1, WORDS_PER_FRAME=3, OUTPUT_ORDER=1
IB
QB
IA
ID
QD
IA
QA
IB
QB
IC
QC
ID
QD
QA
IA
Single Output Stream: NSERIAL=0, WORDS_PER_FRAME=7, OUTPUT_ORDER=0
IB
QB
IC
QC
(c) ONE, TWO OR FOUR CHANNEL MUX MODES
(the timing shown is for SFS_MODE=2,and BLOCK_SIZE=3)
Figure 15. Serial Output Formats
the output page). When the EN_4_FS bit is set, the two bit
tag is used to generate the four frame strobes FSA (tag=0),
FSB (tag=1), FSC (tag=2) and FSD (tag=3). The timing for
these frame strobes are the same as shown for SFS in Figure
15. The separate frame strobe mode (TAG_22=1 and
EN_4_FS=1) is valid for all serial word sizes, but the 2 bit tag
will only be output for 24, 28 or 32 bit serial word sizes. The
four frame strobes share output pins P4, P5, P6 and P7.
3.8.5 Nibble Mode Output
The four serial output pins (P0, P1, P2 and P3), the
serial clock pin (SCK) and the frame sync pin (SFS) can be
configured as a nibble wide output by setting
OUTPUT_MODE=2, MASTER=1 and NIBBLE=1 in address
18. These pins are in a tristate condition when the chip
powers up and need to be enabled by setting EN_SCK,
EN_SFS, EN_P0, EN_P1, EN_P2 and EN_P3 in address 16.
The nibble mode functions the same as the serial mode
except 4 bits are output at a time, the word size is restricted
 GRAYCHIP,INC.
to being 16 or 32 bits, all channels are output on the same
nibble port, and the SFS signal is coincident with the first
nibble of a transfer, rather than being one SCK clock early.
The nibble mode requires that OUTPUT_ORDER=0 and
NSERIAL=0.
For synchronous channels, BLOCK_SIZE should be set
to the number of active channels, WORDS_PER_FRAME
should be set to match the number of active words (two per
active channel for complex data) and FRAME_LENGTH
should be set to define the minimum number of words per
frame.
Asynchronous channels must use BLOCK_SIZE=0 and
TAG_EN=1. WORDS_PER_FRAME must be set to 0 for real
data and 1 for complex data.
The nibble mode can support TDM of multiple chips, in
which case the RDY pin is used to synchronize master and
slave chips in the same way it was used in the serial TDM
mode. To use the TDM mode set MASTER=1 and
- 21 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
EN_RDY=1 in the master chip, and set MASTER=0 and
EN_RDY=0 in the slave chips.
The SFS strobe behaves the same in the nibble mode
as in the serial mode except that the SFS strobe is coincident
with the first nibble in the word and that the SFS strobe can
not come every word in the real 16 bits per word mode
(REAL_ONLY=1,
BITS_PER_WORD=0).
The
latter
exception means that for 16 bit real data, SFS_MODE must
be equal to 0 (one SFS strobe at the beginning of the frame)
AND the frame length cannot be one word
(FRAME_LENGTH can not equal 0). This restriction does not
apply to complex outputs or to 32 bit nibble output modes
The separate frame strobe mode described for
asynchronous serial data can also be used in the nibble
mode.
3.8.6 LINK Mode Output
The four serial output pins (P0, P1, P2 and P3) and the
serial clock (SCK) and RDY pins can be configured as a
nibble wide link port by setting OUTPUT_MODE=2,
NIBBLE=1 and LINK=1 in address 18. The link port can feed
an ADSP-2106x SHARC DSP chip’s link port. The P0, P1,
P2, P3 and SCK pins are in a tristate condition when the chip
powers up and need to be enabled by setting EN_SCK,
EN_P0, EN_P1, EN_P2 and EN_P3 in address 16. EN_RDY
must be low. In the LINK mode the RDY output pin becomes
the ACK (acknowledge) input pin which is tied to the link port
“LACK” signal. LACK is used to stall the output until the
processor is ready for it.
The outputs are transmitted in four bit nibbles on the
rising edge of SCK (INV_SCK in address 17 must be low). If
the ACK signal is low at the end of a 32 bit transfer, then the
clock will remain high and the transmission of the next word
will be delayed until ACK goes high again.
The link port transfers data as 32 bit packets. The user
can choose to transmit two 16 bit words per packet, or a
single 32 bit word. To transmit two 16 bit words per packet
the user must set BITS_PER_WORD to 0 and
WORDS_PER_FRAME to 1. To transmit a single 32 bit word
per packet the user must set BITS_PER_WORD to 1 and
WORDS_PER_FRAME to 0.
The link mode requires OUTPUT_ORDER, NSERIAL
and BLOCK_SIZE to be set to 0. FRAME_LENGTH and
SFS_MODE are unused and should also be set to 0. The link
mode clock rate is set by SCK_RATE.
If the outputs are synchronous, and the chip has been
initialized properly (see Section 3.12), then the first transfer
will be the I part of channel A followed by the Q part, followed
 GRAYCHIP,INC.
by the I/Q pairs from channels B, C and D. If the channels are
asynchronous, or initialization is not possible, then tag bits
must be used to identify the channel data. The tag bits for
synchronous data may be disabled once synchronization is
achieved.
The link mode normally puts the least significant bit in
P0. Note that this is opposite of the GC4014. For pin
compatibility with the GC4014 the control SMUX_0 in
address 22 should be set to 1. This will put the least
significant bit in P3.
When transferring two 16 bit words in a 32 bit link packet
the first word will end up in the upper 16 bits of the packet and
the second word in the lower 16 bits. This means that the
memory order in the DSP chip may end up being (QA, IA,
QB, IB, ...) for complex data and (I1, I0, I3, I2, ...) for single
channel real data. The REVERSE_IQ control bit in address
18 will eliminate this problem by swapping the I/Q pair (or the
I0/I1 pair) in the 32 bit packet.
3.8.7 Parallel Mode Output
The SCK, SFS, RDY and P0 through P23 pins are used
in the parallel mode to output 24 bit wide data samples. The
mode is enabled by setting the EN_SCK, EN_SFS,
EN_RDY, EN_P0, EN_P1, EN_P2, EN_P3 and EN_PAR bits
in address 16, and by setting OUTPUT_MODE=3,
MASTER=1, and PARALLEL=1 in address 18. The parallel
mode also requires NSERIAL=0 and OUTPUT_ORDER=0 in
address 21. FRAME_LENGTH is unused and should be set
to 0 in address 20.
The 24 bit samples are clocked out on the rising edge of
SCK (or the falling edge if INV_SCK is set). The SCK clock
is continuous and the SFS and RDY outputs are used to
identify when a valid sample has been clocked out. The data
valid flags will go high during the SCK clock period when the
24 bit sample is valid. The SFS and RDY flags behave in two
modes as controlled by the SFS_MODE control bits in
address 19.
If SFS_MODE is 0 or 1, then the SFS is IVALID and RDY
is QVALID. If the outputs are synchronous, and the chip has
been initialized properly (see Section 3.12), then the first
valid sample after initialization will be the I part of channel A.
The next valid sample will be the Q part, followed by the I/Q
pairs from channels B, C and D. If the channels are
asynchronous, or initialization is not possible, then tag bits
must be used to identify the channel data. The tag bits for
synchronous data may be disabled once synchronization is
achieved.
- 22 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
SCK
SFS (I Valid)
RDY (Q Valid)
P[0:23]
AI0
AQ0
BI0
BQ0
CI0
CQ0
DI0
DQ0
AI1
AQ1
BI1
AI1
AQ1
BI1
(a) Synchronous Channels, SFS_MODE=0, BLOCK_SIZE=3
SCK
SFS (I or Q Valid)
RDY (Frame Start)
P[0:23]
AI0
AQ0
BI0
BQ0
CI0
CQ0
DI0
DQ0
(b) Synchronous Channels, SFS_MODE=2, BLOCK_SIZE=3
SCK
SFS (I Valid)
RDY (Q Valid)
P[0:23]
AI0
AQ0
BI0
BQ0
CI0
CQ0
(c) Asynchronous Channels, SFS_MODE=0, BLOCK_SIZE=0
SCK
SFS (I or Q Valid)
RDY (Frame Start)
P[0:23]
AI0
AQ0
BI0
BQ0
CI0
CQ0
(d) Asynchronous Channels, SFS_MODE=2, BLOCK_SIZE=0
Figure 16. Parallel Mode Timing
If SFS_MODE is 2 or 3, then SFS is a data valid flag (I
or Q) and the RDY output is a start of frame flag which
identifies the I part of channel A. If the outputs are
synchronous, then the channel data can be identified by its
position in the frame relative to the RDY flag. If the outputs
are asynchronous, then tags must be used to identify the
channel data. The parallel output timing is illustrated in
Figure 16. Note that when SFS_MODE is 2 or 3, then the
RDY flag goes high at the end of the frame and goes low after
the first word of the next frame. This is illustrated in Figures
16 b and d.
The polarities of the SFS and RDY flags are controlled
by the INV_SFS and INV_RDY bits in address 17.
3.9
CLOCKING
The clock rate is equal to the input data rate which can
be up to 80 <TBD 90> MHz. An internal clock doubler
doubles the clock rate so that the internal circuitry is clocked
at twice the data rate. The clock doubler requires a
 GRAYCHIP,INC.
continuous clock (no deleted clock pulses) for proper
operation.
The DVAL input is used as an active low data valid
signal which is clocked into the chip on the rising edge of CK.
The data in the next CK cycle is ignored when DVAL is high.
The DVAL signal operates by gating the output of the clock
doubler. It does not affect the clock to the output circuitry, so
the output will continue while DVAL is high. The DVAL signal
allows users to input data in bursts, such as data which is
being read from a memory or FIFO. The DVAL signal should
never be high for more than a 1msec. Normally DVAL will be
grounded.
The clock doubler can be bypassed and an externally
generated 2X clock can be used in its place by setting the
CK_2X_EN control bit in address 0. In this mode the DVAL
pin becomes the external CK_2X input. This mode is
intended for test purposes only.
The CK_2X_TEST control bit in address 0 enables the
clock test mode where the internal doubled clock is output on
the SO pin. The DVAL pin must be low in this test mode.
- 23 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
3.10
DATA SHEET REV 1.0
POWER DOWN MODES
The 3 bit sync mode control for each sync circuit is
defined in Table 7:
The chip has a power down and clock loss detect circuit.
This circuit detects if the clock is absent long enough to
cause dynamic storage nodes to lose state. If clock loss is
detected, an internal reset state is entered to force the
dynamic nodes to become static. The control registers are
not reset and will retain their values, but any data values
within the chip will be lost. When the clock returns to normal
the chip will automatically return to normal. In the reset state
the chip consumes only a small amount of standby power.
The user can select whether this circuit is in the automatic
clock-loss detect mode, is always on (power down mode), or
is disabled (the clock reset never kicks in) using the
CK_LOSS_DETECT and GLOBAL_RESET control bits in
address 0. The whole chip, or individual down converter
channels can be powered down.
NOTE: Resampler channel 0 provides the “block
complete”
signal
to
the
output
FIFO
when
OUTPUT_ORDER=1 or 2. This means that the channel
feeding resampler channel 0, typically channel A (see
CHAN_MAP in the resampler control page), can not be
powered down when OUTPUT_ORDER=1 or 2.
3.11
SYNCHRONIZATION
Each GC4016 chip can be synchronized through the use
of one of two sync input signals, an internal one shot sync
generator, or a sync counter. The sync to each circuit can
also be set to be always on or always off. Each circuit within
the chip, such as the sine/cosine generators or the
decimation control counter can be synchronized to one of
these sources. These syncs can also be output from the chip
so that multiple chips can be synchronized to the syncs
coming from a designated “master” GC4016 chip.
Table 7: Sync Modes
MODE
SYNC SOURCE
0,1
off (never asserted)
2
SIA
3
SIB
4
ONE_SHOT
5
TC (terminal count of internal counter)
6,7
on (always active)
NOTE: the internal syncs are active high. The SIA and
SIB inputs have been inverted to be the active high syncs SIA
and SIB in Table 7.
The ONE_SHOT can either be a level or a pulse as
selected by the OS_MODE control bit. The level mode is
used to initialize the chip, the pulse mode is used to
synchronously switch frequency, phase or gain values.
Typically the decimation counters (DEC_SYNC), the
flush circuits (FLUSH_SYNC) and the output block sync
(OUT_BLK_SYNC) will be set to SIA, while the NCO phase
accumulator syncs (NCO_SYNC) will be set to SIB. The SIA
input can then be used to initialize and flush the channels and
the SIB sync input can be used, if desired, to synchronize the
phases of the NCOs.
The recommended sync mode settings are summarized
in Table 8.
The SIA and SIB sync inputs are either connected to a
user defined sync generator, for example, an FPGA, or are
tied to a GC4016 chip’s sync output pin (SO). If there are
multiple GC4016 chips in the system, then the SO pin of one
chip can be used to drive the SIA input of all chips, and the
SO pin of another chip can drive the SIB inputs of all chips.
Table 8: Recommended Sync Settings
Global Syncs (Addresses 4 and 5)
Channel Syncs (Pages 7, 15, 23 and 31)
Sync
Value
Description
Sync
Value
Description
DIAG_SYNC
7 (always)
Only used during diagnostics
PHASE_SYNC
7 (always)
Use phase settings as they are loaded
OUTPUT_SYNC
4 (OS)
The SO output is used during initialization
FREQ_SYNC
7 (always)
Use frequency settings as they are loaded
COUNTER_SYNC
4 (OS)
Sync counter with one shot pulse
NCO_SYNC
2 (SIA)
Sync the phase accumulator during
initialization. Set to SIB for frequency
hopping.
DITHER_SYNC
0 (never)
or 2 (SIA)
Can free run except during diagnostics, or
reset during initialization
ZPAD_SYNC
2 (SIA)
Sync the zero pad circuit during initialization
DEC_SYNC
2 (SIA)
Sync the channel decimation during
initialization
Output Circuit Sync (Page 98)
OUT_BLK_SYNC
2 (SIA)
Sync the output block during initialization
Resampler syncs (Page 64)
RES_SYNC
2 (SIA)
Sync the Resampler during initialization
FLUSH_SYNC
2 (SIA)
Flush the channels during initialization
RATIO_SYNC
7(always)
Set to always except when synchronously
changing ratios
GAIN_SYNC
7 (always)
Use fine gain settings as they are loaded
PEAK_SYNC
5 (TC)
Periodically capture peak count data
 GRAYCHIP,INC.
- 24 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
This arrangement allows the user to use the SO sync output
to synchronously drive the SIA or SIB sync inputs of all chips.
The sync source for SO is selected using the
OUTPUT_SYNC control bits in address 4.
(1) Reset the chip by setting address 0, the global reset register,
to 0xF8;
(2) Configure the rest of the chip including setting the
DEC_SYNC, RES_SYNC and OUT_BLK_SYNC to be
ONE_SHOT (mode 4) and the OS_MODE to be 1;
The resampler time delay accumulator is synchronized
by the RES_SYNC control. Typically the resampler is only
synchronized during initialization.
(3) Assert the syncs by setting ONE_SHOT high;
(4) Release the global resets by setting address 0 to 0x08; and
(5) Release the syncs by setting ONE_SHOT to 0.
3.12
INITIALIZATION
3.13
Two initialization procedures are recommended. The
first is recommended for multi-GC4016 chip configuration.
The second can be used for stand alone GC4016 chips.
3.12.1 Initializing Multiple GC4016 Chips
The multi-GC4016 initialization procedure assumes that
the SIA sync input pins of all GC4016 chips are tied together
and are connected to the SO output of the “master” chip, or
to a common sync source. The procedure is to:
(1) Reset the chip by setting address 0, the global reset register,
to 0xF8;
(2) Configure the rest of the chip including setting the
DEC_SYNC, RES_SYNC and OUT_BLK_SYNC to be SIA,
the OS_MODE to be 1, and the OUTPUT_SYNC to be OS
(see Table 7);
(3) Assert the SIA sync input by setting ONE_SHOT high (or by
setting the external SIA source low);
(4) Release the global resets by setting address 0 to 0x08; and
(5) Release the SIA sync by setting ONE_SHOT to 0 (or the
external SIA source high).
The global resets are asserted before configuring the
chip so that the operation of all of the pins, including the
directions of the bidirectional and tristate pins, will be
established before the global resets release them. The SIA
sync is asserted before releasing the global resets so that the
channels will remain in a reset state after the global resets
are released. All channels, the resampler and the output
block will then start synchronously by releasing the SIA sync.
If there are multiple chips which need synchronized, then
synchronously releasing the SIA sync to them all will force
them all to be synchronized.
3.12.2 Initializing Stand Alone GC4016 chips
The initialization sequence for a stand alone GC4016
chip is similar to the one for the multi-GC4016 procedure,
except that the ONE_SHOT is used to synchronize the chip,
not the SIA input sync. The procedure is to:
 GRAYCHIP,INC.
DATA LATENCY
The data latency through the chip is defined as the delay
from the rising edge of a step function input to the chip to the
rising edge of the step function as it leaves the chip. This
delay is dominated by the number of taps in each of the
filters. An estimate of the overall latency through the chip,
expressed as the number of input clock cycles is:
(CIC latency = 2.5N) + (CFIR latency = 0.5N*CTAP) +
(PFIR latency = N*PTAP) + (Resampler latency =
2N*NMULT) + (Output delay) + (Pipeline delay)
where N is the CIC decimation ratio, CTAP is the number of
CFIR taps, and PTAP is the number of PFIR taps. CTAP and
PTAP are normally 21 and 63. Latency can be reduced by
using the NO_SYM_CFIR and NO_SYM_PFIR modes to
shorten these filters. The latency in the resampler can be
minimized by using the bypass configuration (See Section
3.5.6).
The Output delay depends upon the output mode, but is
approximately equal to the fifo block size (BLOCK_SIZE+1)
times the output’s sample period. The Pipeline delay is
approximately 40 clock cycles.
3.14
DIAGNOSTICS
The chip has an internal ramp generator which can be
used in place of the data inputs for diagnostics. An internal
checksum circuit generates a checksum of the output data to
verify the chip’s operation. Section 7.13 gives suggested
checksum configurations and their expected checksums.
Besides the internal diagnostics, the chip can support
board level testing. An output test configuration which can
help initial debug as well as production test is described in
Section 7.14.
3.15
JTAG
The GC4016 supports a four pin (TDI, TDO, TCK and
TMS) boundary scan interface. Contact GRAYCHIP to
receive the GC4016’s BSDL file.
- 25 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
3.16
DATA SHEET REV 1.0
MASK REVISION REGISTER
An 8 bit mask revision code (REVISION) can be read from address 27 of the output control page (page 98). The revision
code allows users to determine, through software, what version of the GC4016 chips are being used. The current mask revision
codes are:
Table 9: Mask Revisions
GC4016
Revision Code
(REVISION)
Release Date
Mask Code on
Package
Description
0
April 2000
SAMPLE
Early samples
1
Sept. 2000
1001ACBA
First Release, Has 40 ohm Vcore to Vpad short, no JTAG
2
March 2001
1001ACBB
Production Release, JTAG added, short removed.
 GRAYCHIP,INC.
- 26 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
4.0
PACKAGING
4.1
GC4016-PB 160 Ball Plastic Ball Grid Array (PBGA)
The GC4016 chip is packaged in a 160 lead plastic ball grid array package.
0.36 mm
A2
1.56 mm
D1
13 mm
A1
A
DIMENSION
D (width body)
D1 (width cover)
P (ball pitch)
B (ball width)
L (overhang)
A (overall height)
A1 (ball height)
A2 (substrate thickness)
0.4 mm
GRAYCHIP
GC4016-PB
QUAD DDC
MMMMMM
LLL YYWW
D
15 mm
1 2 3 4 5 6 7 8 9 10 11 12 13 14
A
B
C
D
E
F
G
H
J
K
L
M
N
P
160 BALL PLASTIC BALL GRID ARRAY
TYP
15 mm
13 mm
1.0 mm
0.5 mm
1.0 mm
1.56 mm
0.4 mm
0.36 mm
TOLERANCE
+-0,05 mm
+0.7 -0.05 mm
+-0.05 mm
+-0.1 mm
+-0.05 mm
+0.19 -0.21 mm
+-0.1 mm
+-0.05 mm
MARKING:
MMMMMM = Mask Code
LLL = Lot Code
YYWW = Date Code
TOP VIEW
L
1.0 mm
P
N
M
L
K
J
H
G
F
E
D
C
B
A
P
VCORE:
C14, D2, L5, L13, M8, P3, P10
VPAD:
B14, C3, D6, D8, D10, G12, M3, P5, P12
GND:
A3, A12, L7, L9, M1, M5, N11
A8, A13, C1, D11, E4, G11, M14, N3, P4
GND (THERMAL):
G7, G8, H7, H8
UNUSED:
A2, B1, B2, B3, B13, C2, M2, M12, M13,
N1, N2, N12, N13, N14, P2, P13
JTAG:
B12 (TCK), C10 (TMS), C12 (TDO), C13 (TDI)
1.0 mm
B
0.5 mm
NOTE: 0.01 to 0.1 µf DECOUPLING CAPACITORS SHOULD BE PLACED
AS CLOSE AS POSSIBLE TO THE MIDDLE OF EACH SIDE OF THE CHIP
1 2 3 4 5 6 7 8 9 10 11 12 13 14
BOTTOM VIEW
Figure 17. GC4016 160 Pin Plastic Ball Grid Array
 GRAYCHIP,INC.
- 27 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Schematic symbols and pin assignments for the GC4016 in each of its input configurations are shown below.
N5
N6
P6
L6
N7
L8
P9
M9
M10
PORT B
L10
(MSB) P23
CIN13 (MSB)
P22
CIN12
CIN11
P21
P20
CIN10
P19
CIN9
P18
H4
P11
P10
CIN1
CIN0
P9
BIN13 (MSB)
BIN12
P8
(FSD) P7
(FSC) P6
BIN11
BIN10
(FSB) P5
(FSA) P4
(SOUTD) P3
(SOUTC) P2
(SOUTB) P1
(SOUTA) P0
(MSB) C7
C6
C5
C4
C3
C2
PORT A
J12
AIN13 (MSB)
H13
AIN12
H14
AIN11
H11
AIN10
H12
AIN9
G14
AIN8
G13
AIN7
F14
AIN6
F12
AIN5
F13
C1
C0
(MSB) A4
A3
AIN4
AIN3
A2
A1
A0
M6
K4
J4
H3
F4
D4
D5
D7
C7
J12
H14
B11
H12
WR
H13
H11
G14
B7
A7
C8
G13
F14
F12
F13
B8
E14
C9
E12
A9
B9
E13
D14
J3
B5
C5
J2
A5
E1
B6
E3
C6
F2
F1
C4
B4
CE
F3
G2
G1
G4
D3
SIA
SO
D1
E2
P8
(MSB) P23
DIN4
P22
BIN13 (MSB)
BIN12
P21
P20
P19
P18
P12
P11
P10
P9
P8
(FSD) P7
(FSC) P6
(FSB) P5
AIN13 (MSB)
AIN12
(FSA) P4
(SOUTD) P3
AIN11
(SOUTC) P2
AIN10
(SOUTB) P1
AIN9
(SOUTA) P0
E11
F11
K11
L11
N10
K2
F3
N9
N8
M7
L3
G1
M6
AIN5
(MSB) C7
C6
C5
AIN4
C4
AIN3
AIN2
C3
C2
AIN1
C1
AIN0
C0
AIN6
DIN1
(MSB) A4
A3
DIN0
A2
DIN13
DIN12
DIN11
DIN10
DIN9
DIN8
DIN7
DIN6
A1
A0
L4
H2
H4
N6
J3
H3
F4
L6
D4
D5
D7
C7
D9
B10
A11
C11
B11
SIA
A6
B7
A7
C8
B8
C9
A9
B9
RD
L8
J12
H14
L10
B5
C5
H12
A5
B6
C6
M11
G13
L14
F12
K14
C4
E14
B4
CE
J13
E13
SO
P8
DVAL
14 Bit Input
Configuration
A10
DIN2
DIN0
CIN0
(MSB) P23
DIN13 (MSB)
P22
CIN13
DIN11
P21
P20
CIN11
P19
DIN9
CIN9
P18
L11
N10
H4
P11
P10
P9
BIN11
C3
C2
C1
C0
AIN11
AIN9
(MSB) A4
A3
BIN7
A2
AIN7
A1
BIN5
A0
BIN9
AIN5
BIN3
AIN3
BIN1
M6
K4
J4
H3
F4
D4
D5
D7
C7
A11
C11
H14
B11
H12
RD
H11
G13
F14
F12
F13
E14
B8
C9
E12
A9
B9
E13
D14
J3
B5
C5
J2
A5
B6
E1
E3
C6
F2
A4
WR
H13
G14
A7
C8
C4
B4
CE
F1
F3
G2
G1
G4
AIN1
SO
E2
D1
P7
P8
DVAL
16 Bit Input
Configuration
RDY
D12
D13
A10
CIN5
CIN4
CIN3
CIN2
CIN1 (EXP2)
CIN0 (EXP1)
(MSB) P23
DIN5 (EXP0)
DIN4 (A/B)
P22
P21
P20
P19
P18
P12
P11
P10
P9
P8
(FSD) P7
(FSC) P6
(FSB) P5
AIN13 (MSB)
AIN12
(FSA) P4
(SOUTD) P3
AIN11
(SOUTC) P2
AIN10
(SOUTB) P1
AIN9
(SOUTA) P0
E11
F11
J11
K11
L11
N10
N9
N8
M7
M6
K4
J4
H4
H3
F4
D4
D5
D7
C7
D9
B10
A11
C11
B11
AIN8
AIN7
AIN5
(MSB) C7
C6
C5
AIN4
C4
AIN3
AIN2
C3
C2
AIN1 (EXP2)
C1
AIN0 (EXP1)
C0
AIN6
DIN1 (EXP0)
DIN0 (A/B)
DIN13
DIN12
DIN11
DIN10
DIN9
DIN8
DIN7
DIN6
(MSB) A4
A3
A2
A1
A0
SIA
A6
B7
A7
C8
B8
C9
A9
B9
B5
C5
A5
B6
C6
A4
RD
WR
C4
B4
CE
D3
SO
E2
D1
SIB
CK
SCK
SFS
CIN7
CIN6
GC4016 P17
QUAD P16
P15
RECEIVER P14
CHIP P13
D9
B10
B7
CIN9
CIN8
BIN13 (MSB)
P9
BIN12
M9
BIN11
M10
BIN10
L10
BIN9
P11
BIN8
M11
BIN7
L12
BIN6
L14
BIN5
K13
BIN4
K14
BIN3
K12
BIN2
J13
BIN1 (EXP2)
J14
BIN0 (EXP1)
H2
DIN3 (EXP0)
J1
DIN2 (A/B)
J12
A6
CIN11
CIN10
L8
K11
CIN3
P12
AIN13
L4
N5
N6
H1
J11
N9
N8
M7
BIN13 (MSB)
N4
M4
N7
G3
E11
F11
GC4016 P17
QUAD P16
P15
DIN5 RECEIVER P14
CIN5
CHIP P13
DIN3
DIN7
CIN7
DIN1
CIN1
L3
L2
P6
L6
CIN2
SIA
SIB
P7
RDY
D12
D13
DIN4
CIN4
D3
E2
D1
CK
CIN6
P9
P8
BIN12 (MSB)
H13
(FSD) P7
AIN12
M10
(FSC) P6
BIN10
H11
(FSB) P5
AIN10
P11
(FSA) P4
BIN8
G14
(SOUTD) P3
AIN8
(SOUTC) P2
L12
BIN6
F14
(SOUTB) P1
AIN6
K13
(SOUTA) P0
BIN4
F13
AIN4
K12
BIN2
(MSB) C7
E12
AIN2
C6
J14
C5
BIN0
D14
C4
AIN0
M9
A4
WR
N4
G3
K4
J4
AIN8
AIN7
E1
K1
F2
J11
DIN6
SCK
SFS
PORT C
PLUS
N7
CIN0
DIN5
D3
SIB
P7
P6
J2
GC4016 P17
QUAD P16
P15
RECEIVER P14
CHIP P13
A11
C11
A6
J1
CIN2
CIN1
BIN11
M10
BIN10
L10
BIN9
P11
BIN8
M11
BIN7
L12
BIN6
L14
BIN5
K13
BIN4
K14
BIN3
K12
BIN2
J13
BIN1
J14
BIN0
H2
DIN3
J1
DIN2
D9
B10
A4
RD
E12
AIN2
E13
AIN1
D14
AIN0
P9
M9
L11
N10
CIN2
P12
L8
K11
N9
N8
M7
BIN9
P11
BIN8
M11
BIN7
L12
BIN6
L14
BIN5
K13
BIN4
K14
BIN3
K12
BIN2
J13
BIN1
J14
BIN0
E14
J11
GC4016 P17
QUAD P16
CIN6
P15
CIN5 RECEIVER P14
CIN4
CHIP P13
CIN3
CIN8
CIN7
G3
H1
E11
F11
N5
CIN4
CIN3
DIN8
CIN8
Pins shared with serial outputs
M4
L4
DIN0
M4
H1
CIN10
CIN13 (MSB)
CIN12
(Must be Grounded)
N4
DIN2
DIN1
A10
K2
L1
PORT B
PORT C
L3
L2
P6
L6
N7
RDY
G4
CIN12
DIN10
PORT A
L1
N5
N6
DIN4
DIN3
CIN6
CIN5
D13
G2
L2
K1
K3
DIN12 (MSB)
UNUSED
K3
K2
M4
L4
D12
PORT B
K1
A10
SCK
SFS
MINUS
J2
RDY
L1
CIN9
CIN8
CIN7
PLUS
J3
DIN6
DIN5
L2
N4
CIN11
CIN10
PORT A
H1
H2
J1
DIN7
D12
D13
F1
MINUS
G4
G3
SCK
SFS
E3
K3
CIN13 (MSB)
CIN12
(Must be Grounded)
G1
DIN9
DIN8
PORT C
G2
L1
L3
DIN10
PORT B
PORT D
F1
F3
K3
K2
DIN12
DIN11
PORT A
F2
K1
DIN13 (MSB)
UNUSED
E1
E3
CK
DVAL
14 Bit Differential Input
Configuration
SIB
P7
P8
CK
DVAL
12 Bit Plus Exponent
Configuration
NOTE: All unused inputs must be grounded
JTAG Pins are: B12 (TCK), C10 (TMS), C12 (TDO), C13 (TDI)
Figure 18. GC4016 Pin Assignments
 GRAYCHIP,INC.
- 28 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Table 10: GC4016 Pin Out Locations Top View
1:
2:
A:
3:
4:
5:
6:
7:
8:
9:
10:
11:
12:
13:
14:
*
GND
RD
A2
C7
C5
GND
C1
RDY
P2
GND
GND
B:
*
*
*
CE
A4
A1
C6
C3
C0
P3
P0
TCK
*
VPAD
C:
GND
*
VPAD
WR
A3
A0
P5
C4
C2
TMS
P1
TDO
TDI
VCORE
D:
SIB
VCORE
SIA
P8
P7
VPAD
P6
VPAD
P4
VPAD
GND
SCK
SFS
AIN0
E:
DIN13
SO
DIN12
GND
P23
AIN2
AIN1
AIN3
F:
DIN10
DIN11
DIN9
P9
P22
AIN5
AIN4
AIN6
G:
DIN7
DIN8
DIN5
DIN6
TGND
TGND
GND
VPAD
AIN7
AIN8
H:
DIN4
DIN3
P10
P11
TGND
TGND
AIN10
AIN9
AIN12
AIN11
J:
DIN2
DIN0
DIN1
P12
P21
AIN13
BIN1
BIN0
K:
CIN13
CIN11
CIN12
P13
P20
BIN2
BIN4
BIN3
L:
CIN10
CIN8
CIN9
CIN5
VCORE
CIN1
GND
BIN13
GND
BIN9
P19
BIN6
VCORE
BIN5
M:
GND
*
VPAD
CIN6
GND
P14
P15
VCORE
BIN11
BIN10
BIN7
*
*
GND
N:
*
*
GND
CIN7
CIN4
CIN3
CIN0
P16
P17
P18
GND
*
*
*
*
VCORE
GND
VPAD
CIN2
CK
DVAL
BIN12
VCORE
BIN8
VPAD
*
P:
VPAD =Pad Ring Power (3.3v)
VCORE = Core Power (2.5v)
TGND = Thermal Ground
* = unused ball
 GRAYCHIP,INC.
- 29 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
SIGNAL
DESCRIPTION
A,B,C, DIN
INPUT DATA, Active high input pins
The 14 bit 2’s complement input data for the four channels. The inputs are clocked into the chip on the rising edge of the clock
(CK). These pins can be configured to support three 16 bit ports, two 14 bit differential ports, three 12 bit + 3 bit exponent ports, or
three 12 bit + 3 bit exponent + A/B selection ports. See Section 3.2 and Figure 18.
P[0:3]
BIT SERIAL AND NIBBLE OUTPUT DATA, Active high tristate output pins
The nibble and bit serial output pins. In the serial mode these are individual outputs, in the nibble mode these form a four bit nibble
(P0 is normally the LSB of the Nibble, P3 is the MSB). The output bits are clocked out coincident with the rising edge of SCK (falling
edge if INV_SCK=1). These pins are tristated at power up and are enabled by EN_P0, EN_P1, EN_P2 and EN_P3.
P[0:23]
PARALLEL OUTPUT DATA, Active high tri state output pins
The 24 bit parallel output port. These output bits are clocked out by CK coincident with the rising edge of SCK (falling edge if
INV_SCK=1) with the SFS and RDY pins used to identify valid data (Section 3.8.6). These pins are tristated at power up and are
enabled by the EN_PAR control register bit. These pins can are used in the wide word microprocessor mode. In this mode the pins
are used as part of the control bus when read from the data output page (See Section 3.8).
SCK
SERIAL DATA CLOCK, Active high or low tristate output pin
The serial, nibble, link and parallel data output clock. The SCK signal is clocked out on the rising edge of CK. The SFS, RDY and
P output signals are clocked out of the chip coincident with the active edge of this clock. The active edge of the clock is user
programmable. This pin is tristated at power up and is enabled by the EN_SCK control register bit.
SFS
SERIAL FRAME STROBE, Active high or low tristate output pin
The bit serial word strobe. This strobe identifies the beginning of a frame, a complex pair, or a word within bit serial output streams
as controlled by the SFS_MODE control register bits. The polarity of this signal is user programmable. This pin is tristated at power
up and is enabled by the EN_SFS control register bit. This pin is also used as a data valid signal for parallel outputs.
RDY
READY FLAG, programmable active high or low bidirectional I/O pin
Used to identify when new outputs are available in the serial, nibble and microprocessor output modes. In the link mode RDY is an
input signal tied to the LACK output signal from SHARC DSP chips. In the parallel mode it is a data valid flag. The RDY signal is
clocked out on the rising edge of CK. This pin is tristated at power up and is enabled by the EN_RDY control register bit.
CK
INPUT CLOCK. Active high input pin
The clock input to the chip. The AIN, BIN, CIN, DIN, DVAL, SIA and SIB input signals are clocked into the chip on the rising edge
of this clock. The SO, P, SFS, SCK and RDY outputs are clocked out on the rising edge of this clock.
DVAL
DATA VALID. Active low input pin
This pin is normally grounded. This pin must be low to enable the internal clock. DVAL is clocked into the chip on the rising edge
of CK, and, if high, disables the following CK edge to the channels and resampler. It does not effect the clock to the output circuitry.
Since DVAL enables or disables the internal clock, it can be used as a data enable for non-continuous input data. This pin should
never be held high for more than 1ms. DVAL is used as a 2X clock input when the CK_2X_EN control bit is high (see Section 3.9).
SIA, SIB
SYNC IN A and B. Active low input pins
The sync inputs to the chip. All timers, accumulators, and control counters (except the resampler time delay accumulator) are, or
can be, synchronized to SIA or SIB. These syncs are clocked into the chip on the rising edge of the input clock (CK).
SO
SYNC OUT. Active low output pin
This signal is either a delayed version of one of the input syncs SIA or SIB, the sync counter’s terminal count (TC), or a one-shot
strobe. The SO signal is clocked out of the chip on the rising edge of the input clock (CK).
C[0:7]
CONTROL DATA I/O BUS. Active high bidirectional I/O pins
This is the 8 bit control data I/O bus. Control registers are written to or read from through these pins. The chip drives these pins
when CE is low, RD is low and WR is high. Note that when the output is in the wide word microprocessor mode, the P[0:23] pins
are used when reading and will behave the same as the C[0:7] pins. When reading from the output page the P[0:23] pins will output
data, when reading from all other pages the P[0:23] pins will read back high.
A[0:4]
CONTROL ADDRESS BUS. Active high input pins
These pins are used to address the control registers within the chip. Each of the control registers within the chip are assigned a
unique address. A control register can be written to or read from by setting A[0:4] to the register’s address and setting the page
register appropriately. An alternate method using a 4 bit addressing scheme is available to support using buses with limited
addressing (for example the TI320C6X family peripheral host interface bus). See Section 3.1.
RD
READ ENABLE. Active low input pin
The register selected by A[0:4] and the page register is output on the C[0:7] pins when RD and CE are low.
WR
WRITE ENABLE. Active low input pin
The value on the C[0:7] pins is written into the register selected by the A[0:4] and page register when WR and CE are low.
CE
CHIP ENABLE. Active low input pin
This control strobe enables the read or write operations.
TCK,TDI,TMS,TDO
JTAG INTERFACE. Active high input (TCK, TMS, TDI) and tristate output (TDO) pins
The JTAG interface (See Section 3.15)
 GRAYCHIP,INC.
- 30 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.0
DATA SHEET REV 1.0
CONTROL REGISTERS
The chip is configured by writing to eight bit control registers. These registers are accessed for reading or writing using the
control bus pins (CE, RD, WR, A[0:4], and C[0:7]) described in Section 3.1. The 32 word address space is split into eight global
registers (addresses 0-7), eight unused registers (addresses 8-15) and 16 paged registers (addresses 16-31). The global
registers are available from each page. Address 2 is the page register which selects which control registers are accessed by
addresses 16 through 31.
5.1
GLOBAL CONTROLS
The eight global control registers are:
Table 11: Global Control Registers
ADDRESS
NAME
DESCRIPTION
0
Global Reset
Miscellaneous general controls.
1
Status
uP ready/missed, resampler overflows, checksum ready,
2
Page
Page register for both 4 and 5 bit addressing modes
3
Checksum
Checksum results register
4
General Syncs
Syncs for output, checksum. One shot.
5
Count Sync
Sync for counter, ramp selection
6
Counter Byte 0
Ramp counter least significant byte.
7
Counter Byte 1
Ramp counter most significant byte.
ADDRESS 0:
Global Reset, Powers up as 0xF0. Suggested Default is 0xF8 during configuration,
0x08 afterwards.
BIT
TYPE
NAME
DESCRIPTION
0 LSB
R/W
CK_LOSS_DETECT
Disable the clock loss detection circuitry. This circuitry protects the chip against
a current surge that may result if the clock is inactive for more than 100mS.
Setting this bit turns off the clock loss detection circuitry. Used for test, not
recommended for general use.
1
R/W
CK_2X_TEST
Test mode to output the doubled clock on SO. DVAL must be low. Normally set to
zero.
2
R/W
CK_2X_EN
Changes the DVAL pin to become a CK_2X input pin. The chip normally uses
an internally generated doubled clock (twice the CK clock). Setting this bit allows
an externally generated doubled clock to be used. Used in test, not
recommended for general use.
3
R/W
EDGE_WRITE
Sets the edge write mode for the control interface. When low the data must be
stable while the write strobe is low. When high the outputs are latched on the
rising edge of the write strobe (WR | CE). A short (approximately 15 nS) write
recovery time is required during which time the chip should not be read from or
written to. Recommended to be set high.
4
R/W
RESAMPLER_RESET
5
R/W
PAD_RESET
Forces SFS and RDY pads to tristate during power up. The user needs to clear
this bit for proper operation of SFS and RDY.
6
R/W
OUT_BLK_RESET
This bit resets the output formatter block. This bit is set during power up and is
cleared after configuration.
7
R/W
GLOBAL_RESET
This bit powers down the chip and tristates the output pins. This bit is set during
power up and is cleared after configuration.
 GRAYCHIP,INC.
This bit resets the resampler. This bit is set during power up and is cleared after
configuration.
- 31 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 1:
DATA SHEET REV 1.0
Status Register, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0 LSB
R/W
READY
The user sets this bit after reading the output registers. The chip clears this bit
when new values have been loaded and it is time to read them.
1
R/W
MISSED
The chip sets this bit If the user has not set the READY bit before the chip loads
the output registers. This bit high indicates that an error has occurred.
2
R/W
RES_IOV
This control bit is set by the chip whenever the resampler I channel overflows.
The user can monitor this bit to see if the gain is too high. The user clears the bit
by writing a zero to it.
3
R/W
RES_QOV
This control bit is set by the chip whenever the resampler Q channel overflows.
The user can monitor this bit to see if the gain is too high. The user clears the bit
by writing a zero to it.
4
R/W
CHECK_DONE
This bit is set when the checksum sync is active (see DIAG_SYNC in address
4). The user can count sync cycles by clearing this bit and then waiting for it to
be set. The checksum will be complete after it this bit has been cleared and set
four times.
5-7 MSB
R
ZERO
Reads back as zero.
The READY bit is used to tell an external processor when new output samples are ready to be read. If desired, the RDY pin
can be used as an interrupt to the external processor (See Section 3.8.1) to tell the processor when to read new samples. The
user does not need to set the READY bit if RDY is used. If READY is not set, however, the MISSED flag will not be valid.
ADDRESS 2:
Page Register
BIT
TYPE
NAME
DESCRIPTION
0 LSB
R/W
A3
Used in the 4_BIT_ADDRESS mode (see global register 4) as address bit A3.
This bit is unused if 4_BIT_ADDRESS=0.
1-7 MSB
R/W
PAGE[0:6]
Page number for addressing different portions of the chip.
In normal mode, the LSB is unused. PAGE is decoded to select the different control pages (see Table 12). Note that
addresses 0-7 are globally visible regardless of the PAGE value. Addresses 8-15 are unused. Addresses 16-31 are paged. In
the four bit address mode, the LSB provides address bit A3, effectively reducing each page to 8 words and doubling the number
of pages.
ADDRESS 3:
Checksum Register
BIT
TYPE
NAME
DESCRIPTION
0-7
R
CHECKSUM[0:7]
The checksum.
The checksum register is a read only register which contains the checksum of the output data. The checksum is stored in
the checksum register and then starts over again each time the DIAG_SYNC (See address 4) occurs.
 GRAYCHIP,INC.
- 32 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 4:
DATA SHEET REV 1.0
General Sync Register, Suggested default = 0x27, Cleared by power up
BIT
TYPE
NAME
DESCRIPTION
0-2
(LSB)
R/W
DIAG_SYNC
The Checksum generator is strobed by this sync. See Table 7 for the possible
sync selections.
3-5
R/W
OUTPUT_SYNC
The selected sync is inverted and output on the SO pin. See Table 7.
6
R/W
4_BIT_ADDRESS
This mode allows four address bits (such as the expansion bus of the
TI320C6202) instead of 5 to be used. In this mode each page contains 8 words
(rather than 16). The LSB of the page register is used as address bit A3. Pin A3
should be grounded in this mode.
7 (MSB)
R/W
DIFF_IN
Enables differential receivers. Both this bit and the corresponding bits in each
channel must be set to properly receive differential signals. This bit should be
cleared for other inputs and for minimum power consumption.
Contact Graychip for the use of the differential input mode.
ADDRESS 5:
Count Sync Register, Suggested default = 0x50
BIT
TYPE
NAME
DESCRIPTION
0-1 LSB
R/W
DIAG_SOURCE
This two bit field selects the diagnostic input source used when the INPUT_SEL
field in each channel’s control register is set to 6 or 7 (See channel control
address 27). DIAG_SOURCE = 0 selects the 16 LSB’s of the counter (see
addresses 6&7) as a diagnostic ramp. DIAG_SOURCE=1 is a zero input,
DIAG_SOURCE=2 is unused, DIAG_SOURCE= 3 gives a 0x4000 constant
input.
2-4
R/W
COUNTER_SYNC
Synchronizes the sync counter. This counter is used to generate the periodic TC
sync. See Table 7.
5
R/W
COUNT_TEST
Used during factory tests. Should be set to 0 for normal operation.
6
R/W
OS_MODE
The ONE_SHOT signal is a level, not a pulse when this bit is set.
7 (MSB)
R/W
ONE_SHOT
The one shot sync signal (OS) is generated when this bit is set. If OS_MODE is
low, then a one shot pulse (one clock cycle wide) is generated. If OS_MODE is
high, then the ONE_SHOT sync is active while this bit is high. This bit must be
cleared before another one shot pulse can be generated.
ADDRESS 6:
Counter Byte 0, Suggested default = DEC (CIC decimation value)
BIT
TYPE
NAME
DESCRIPTION
0-7
R/W
CNT[0:7]
The LSBs of the counter cycle period
ADDRESS 7:
Counter Byte 1, Suggested default = DEC
BIT
TYPE
NAME
DESCRIPTION
0-7
R/W
CNT[8:15]
The 8 MSBs of the counter cycle period
The chip’s internal sync counter counts in cycles of 256(CNT+1) clocks. A terminal count signal (TC) is output at the end of
each cycle. The counter can be synchronized to an external sync as specified in the Count Sync Register (address 5). If CNT is
set so that 256(CNT+1) is a multiple of sixteen times the CIC decimation ratio (i.e., a multiple of 16N), then the terminal count of
this counter can be output on the SO pin and used to periodically synchronize multiple GC4016 chips.
 GRAYCHIP,INC.
- 33 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.2
DATA SHEET REV 1.0
PAGED REGISTERS
Addresses 16 to 31 are used in pages as determined by the page map register (address 2).
Table 12: Page Assignments
Page
Description
0,1
Channel A CFIR Coefficients
2-5
Channel A PFIR Coefficients
6
Channel A Frequency
7
Channel A Control
8,9
Channel B CFIR Coefficients
10-13
Channel B PFIR Coefficients
14
Channel B Frequency
15
Channel B Control
16,17
Channel C CFIR Coefficients
18-21
Channel C PFIR Coefficients
22
Channel C Frequency
23
Channel C Control
24,25
Channel D CFIR Coefficients
26-29
Channel D PFIR Coefficients
30
Channel D Frequency
31
Channel D Control
32-63
Resampler Coefficients
64
Resampler Control
65
Resampler Ratios
66-95
unused
96,97
Output Data
98
Output Control
99-127
unused
The following sections describe each of these pages.
 GRAYCHIP,INC.
- 34 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.3
DATA SHEET REV 1.0
CFIR COEFFICIENT PAGES
The user programmable filter CFIR coefficients are stored using pages 0 and 1 for channel A, pages 8 and 9 for channel B,
pages 16 and 17 for channel C, and pages 24 and 25 for channel D.
Table 13: CFIR Coefficient Pages
Pages 0, 8, 16, or 24
Pages 1, 9, 17, or 25
Address
Description
Description
16
h0 LSBs (end or first tap)
h8 LSBs
17
h0 MSBs (end or first tap)
h8 MSBs
18
h1 LSBs
h9 LSBs
19
h1 MSBs
h9 MSBs
20
h2 LSBs
h10 LSBs (center tap)
21
h2 MSBs
h10 MSBs (center tap)
22
h3 LSBs
unused
23
h3 MSBs
24
h4 LSBs
25
h4 MSBs
26
h5 LSBs
27
h5 MSBs
28
h6 LSBs
29
h6 MSBs
30
h7 LSBs
31
h7 MSBs
Coefficient h0 is the first coefficient and coefficient h10 is the center coefficient of the filter’s impulse response. The 16 bit 2’s
complement coefficients are stored in two bytes, least significant byte first, for example, the LSBs of coefficient 0 are stored in
address 16 and the MSBs in address 17.
TO LOAD A COEFFICIENT THE USER MUST WRITE THE LSBYTE FIRST FOLLOWED BY THE MSBYTE. Unknown
values will be written into the LSBs if the MSB is written first. The coefficient registers are read/write.
5.4
PFIR COEFFICIENT PAGES
The user programmable filter PFIR coefficients are stored using pages 2, 3, 4 and 5 for channel A, pages 10, 11, 12 and 13
for channel B, pages 18, 19, 20 and 21 for channel C, and pages 26, 27, 28 and 29 for channel D.
Table 14: PFIR Coefficient Pages
Pages 2, 10, 18, or 26
Pages 3, 11, 19, or 27
Pages 4, 12, 20, or 28
Pages 5, 13, 21, or 29
Address
Description
Description
Description
Description
16
h0 LSBs (end tap)
h8 LSBs
h16 LSBs
h24 LSBs
17
h0 MSBs (end tap)
h8 MSBs
h16 MSBs
h24 MSBs
18
h1 LSBs
h9 LSBs
h17 LSBs
h25 LSBs
19
h1 MSBs
h9 MSBs
h17 MSBs
h25 MSBs
20
h2 LSBs
h10 LSBs
h18 LSBs
h26 LSBs
21
h2 MSBs
h10 MSBs
h18 MSBs
h26 MSBs
22
h3 LSBs
h11 LSBs
h19 LSBs
h27 LSBs
23
h3 MSBs
h11 MSBs
h19 MSBs
h27 MSBs
24
h4 LSBs
h12 LSBs
h20 LSBs
h28 LSBs
25
h4 MSBs
h12 MSBs
h20 MSBs
h28 MSBs
26
h5 LSBs
h13 LSBs
h21 LSBs
h29 LSBs
27
h5 MSBs
h13 MSBs
h21 MSBs
h29 MSBs
28
h6 LSBs
h14 LSBs
h22 LSBs
h30 LSBs
29
h6 MSBs
h14 MSBs
h22 MSBs
h30 MSBs
30
h7 LSBs
h15 LSBs
h23 LSBs
h31 LSBs (center tap)
31
h7 MSBs
h15 MSBs
h23 MSBs
h31 MSBs (center tap)
 GRAYCHIP,INC.
- 35 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Coefficient h0 is the first coefficient and coefficient h31 is the center coefficient of the filter’s impulse response. The 16 bit 2’s
complement coefficients are stored in two bytes, least significant byte first, for example, the LSBs of coefficient 0 are stored in
address 16 and the MSBs in address 17.
TO LOAD A COEFFICIENT THE USER MUST WRITE THE LSBYTE FIRST FOLLOWED BY THE MSBYTE. Unknown
values will be written into the LSBs if the MSB is written first. The coefficient registers are read/write.
5.5
CHANNEL FREQUENCY PAGES
Pages 6, 14, 22, and 30 contain the phase and frequency control settings for the four channels. The frequency and phase
for channel A are set in page 6. The frequency and phase for channel B are set in page 146. The frequency and phase for channel
C are set in page 22. The frequency and phase for channel D are set in page 30. All registers are read/write.
ADDRESSES 16, 17: Phase
ADDRESS
TYPE
NAME
DESCRIPTION
16
R/W
PHASE[0:7]
Byte 0 (LSBs) of PHASE
17
R/W
PHASE[8:15]
Byte 1 (MSBs) of PHASE
The 16 bit phase offset is defined as:
PHASE = 216P/2π
where P is the desired phase in radian from 0 to 2π.
ADDRESSES 18, 19, 20, and 21: Frequency
ADDRESS
TYPE
NAME
DESCRIPTION
18
R/W
FREQ[0:7]
Byte 0 (LSBs) of FREQ
19
R/W
FREQ[8:15]
Byte 1 of FREQ
20
R/W
FREQ[16:23]
Byte 2 of FREQ
21
R/W
FREQ[24:31]
Byte 3 (MSBs) of FREQ
The 32 bit frequency control word is defined as:
FREQ = 2 32F/FCK
where F is the desired tuning frequency and FCK is the chip’s clock rate (CK). Use positive frequency values to downconvert
signals. Use negative frequency values to invert the signal’s spectrum. The 32 bit 2’s complement frequency words are entered
as four bytes, the least significant byte in the lowest address, the most significant in the highest address.
 GRAYCHIP,INC.
- 36 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.6
DATA SHEET REV 1.0
CHANNEL CONTROL PAGES
These pages contain the various control settings for the four channels. To configure channels A, B, C and D use pages 7,
15, 23, and 31 respectfully. All registers are read/write. The following table summarizes the registers:
Table 15: Channel Control Registers
ADDRESS
NAME
DESCRIPTION
16
Channel Reset
Resets the channel and sets SHIFT
17
Frequency Sync
Phase and Frequency syncs
18
NCO sync
NCO and dither syncs
19
Blank
Blank controls
20
Dec Sync
Synchronizes CIC decimation and flush
21,22
Decimation Ratio
Sets CIC decimation and sync for fine gain
23
CIC Scale
Sets CIC gain and MIX20B
24
SplitIQ
SplitIQ and negate control
25
CFIR
Quarter delays, Coarse, Nosym1
26
PFIR
Half delays, Nosym2
27
Input
Selects input format and port.
28
Peak Control
Controls peak and overflow detection counter
29
Peak Read
Number of peaks (overflows) during last sync period.
30,31
Fine Gain
Fine gain control
ADDRESS 16:
Channel Reset Set to 0x80 on power up, Suggested default = 0x0C
BIT
TYPE
NAME
DESCRIPTION
0-2 (LSBs)
R/W
SHIFT
Value used to shift up the mixer output if USE_SHIFT is high. Note that if USE_SHIFT is high
and MIX20B is set, then SHIFT is restricted to the range of 4-7.
3
R/W
USE_SHIFT
Selects SHIFT if high. If low the mixer output shift value is provided by the exponent bits of
the floating point input format (see Section 3.2).
4-6
R/W
Unused
7
R/W
CH_RESET
ADDRESS 17:
This bit resets the channel. If high, the channel is in a reset state with the clock disabled.
Frequency Sync, Suggested default = 0x77
BIT
TYPE
NAME
DESCRIPTION
0-2 LSB
R/W
PHASE_SYNC
The new phase offset takes effect on this sync
3
R/W
Unused
4-6
R/W
FREQ_SYNC
7 MSB
R/W
Unused
The new frequency setting takes effect on this sync.
These syncs use the selections shown in Table 7.
The FREQ_SYNC and PHASE_SYNC are typically set to be “always” so that frequency and phase settings will take effect
immediately as they are written into their control registers. Due to the asynchronous nature of the interface in this mode a few
samples will be mixed with transient frequencies until the new frequency and phase values are established. If these transients
are undesirable OR if the accumulated phase must be controlled, then another sync source should be used.
 GRAYCHIP,INC.
- 37 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 18:
DATA SHEET REV 1.0
NCO Sync, Suggested default = 0x22
BIT
TYPE
NAME
DESCRIPTION
0-2 LSB
R/W
NCO_SYNC
The NCO is initialized to the phase setting by this sync
3
R/W
Unused
4-6
R/W
DITHER_SYNC
7 MSB
R/W
Unused
The dither circuit is reset by this sync to zero.
These syncs use the selections shown in Table 7.
The NCO_SYNC can be changed to “SIB” so that the SIB sync can be used to synchronize the NCO phases of multiple
channels or chips.
The DITHER_SYNC is used to turn on or off the dithering of the NCO phase. Dithering is turned on by setting
DITHER_SYNC to “never”. Dithering is turned off by setting DITHER_SYNC to “always” which causes it to remain reset to zero.
During diagnostics the NCO_SYNC and DITHER_SYNC should be set to “TC” (5).
If the user wishes to allow the chip to free run, asynchronous to other chips, then these sync settings can be set to “never”.
ADDRESS 19:
Zero Pad Mode Control Register, Suggested default = 0x20
BIT
TYPE
NAME
DESCRIPTION
0-3 LSB
R/W
NZEROS
The number of zeroes to insert between each sample in the blank mode.
Ranges from 0 to 15. Note that this introduces a gain of 1/(NZEROS+1).
4-6
R/W
ZPAD_SYNC
The sync selection from Table 7 for the zero pad function (See Figure 5).
7 MSB
R/W
ZPAD_EN
Turn on zero padding for this channel.
ADDRESS 20:
Dec and Flush Sync Register, Suggested default = 0x22
BIT
TYPE
NAME
DESCRIPTION
0-2 LSB
R/W
DEC_SYNC
Synchronizes the decimation control counter. The decimation counter controls
the filtering of each channel.
3
R/W
Unused
4-6
R/W
FLUSH_SYNC
7 MSB
R/W
Unused
Synchronizes the flush sync for this channel.
The decimation counter sync (DEC_SYNC) may be periodic without interrupting processing so long a the sync period is a
multiple of 8N. A decimation sync should be issued if the decimation control is changed.
Each channel should be flushed (FLUSH_SYNC) when the chip is being initialized or when the decimation control is
changed. The flush lasts for 24N clocks after the sync occurs. The channel flush syncs will normally be left in a “never” mode.
During diagnostics the channels will need to be flushed at the beginning of each sync cycle.
The user may wish to flush a channel when a new frequency is selected in order to purge the datapath of the last signal.
 GRAYCHIP,INC.
- 38 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 21:
DATA SHEET REV 1.0
Decimation Ratio Byte 0, Suggested default = 0x07
BIT
TYPE
NAME
DESCRIPTION
0-7
R/W
DEC[0:7]
The LSBs of the decimation control
ADDRESS 22:
Decimation Ratio Byte 1, Suggested default = 0x70
BIT
TYPE
NAME
DESCRIPTION
0-3
R/W
DEC[8:11]
The 4 MSBs of the decimation control
4-6
R/W
GAIN_SYNC
The new fine gain takes effect on this sync.
7 MSB
R/W
Unused
The CIC decimation is N. When processing in normal mode (not SPLITIQ) DEC should be set to N-1 where N ranges from
8 to 4096. When processing in SPLITIQ mode DEC should be set to 2N-1 where N ranges from 4 to 2048.
ADDRESS 23:
CIC Scale, Suggested default = 0x79
BIT
TYPE
NAME
DESCRIPTION
0-2
R/W
SCALE
SCALE ranges from 0 to 5.
3-5
R/W
BIG_SCALE
BIG_SCALE ranges from 0 to 7.
6
R/W
MIX20B
Round mixer output to 20 bits if MIX20B=1 or to 16 bits if MIX20B=0. If
MIX20B=1 then SHIFT (see address 16) must be at least 4 to shift the extra 4
bottom bits into the datapath. For very large decimations (>3104) unity gain
through the CIC can only be accomplished by using a smaller SHIFT and hence
MIX20B must be zero. The default is MIX20B=1.
7
R/W
unused
The CIC filter has a gain which is equal to N5. To remove this gain the CIC inputs are pre-scaled down by
(62-SHIFT-SCALE-6*BIG_SCALE) bits before filtering. The values of SHIFT, SCALE and BIG_SCALE must be such that:
( SHIFT + SCALE + 6 × BIG_SCALE ) ≤ ( 62 – 5log2 N + log2 (NZERO+1) )
Overflows due to improper gain settings will go undetected if this relationship is violated. For example, this restriction means
that for N equal to 8, and SHIFT equal to 4, BIG_SCALE and SCALE should be less than or equal to 7 and 1 respectively.
ADDRESS 24:
SplitIQ, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-3
R/W
NEG_CTL
A four bit pattern that multiplies the output of PFIR by +/- 1. This is used to
perform complex to real conversion (Fs/4 upconvert), flip the spectrum, or
perform an Fs/2 frequency shift.
4
R/W
SPLITIQ
Process only the real data (at twice the throughput). Used to allow two channels
to be ganged together for wider output bandwidth.
5
R/W
IONLY
All samples output by this channel are real. Used in the SPLITIQ mode to
identify the channel generating the I-half of the complex output.
6
R/W
QONLY
All samples output by this channel are imaginary. Used in the SPLITIQ mode to
identify the channel generating the I-half of the complex output. When
performing complex to real conversion this bit must be set (amongst others).
7
R/W
Unused
Sections 3.3 and 3.4 describe how to use these controls.
 GRAYCHIP,INC.
- 39 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 25:
DATA SHEET REV 1.0
CFIR, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0 LSB
R/W
NO_SYM_CFIR
When this bit is high, CFIR is an 11 tap asymmetric filter. Coefficient h0 is
multiplied by the newest data.
1
R/W
QDLY_CFIR
Delay the Q sample stream by one CFIR input sample. Effectively, this is a 1/4
sample delay of the output PFIR. Used in the multi-channel modes (see Section
3.4).
2
R/W
IDLY_CFIR
Delay the I-sample stream by one CFIR input sample. Effectively, this is a 1/4
sample delay of the output PFIR.Used in the multi-channel modes (see Section
3.4).
3
R/W
Unused
4-6
R/W
COARSE
This introduces a gain of 2COARSE after the CIC but before rounding to 20 bits to
feed CFIR. This allows optimization of gain in applications where it is known that
most of the signal energy presented to the A/D has been filtered out by the CIC. An
example of such an application is an FDM stack.
7
R/W
TEST
Test mode. Should be set to 0.
ADDRESS 26:
PFIR, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0 LSB
R/W
NO_SYM_PFIR
When this bit is high PFIR is an 32 tap asymmetric filter. Coefficient h0 is multiplied
by the newest data.
1
R/W
QDLY_PFIR
Delay the Q sample stream by one PFIR input sample. Effectively, this is a 1/2
sample delay of the output PFIR. Used in the real or multi-channel modes (see
Sections 3.3 and 3.4).
2
R/W
IDLY_PFIR
Delay the I-sample stream by one PFIR input sample. Effectively, this is a 1/2
sample delay of the output PFIR.Used in the real or multi-channel modes (see
Sections 3.3 and 3.4).
3-4
R/W
PEAK_SELECT
Select the overflow detection source for the peak counter. Only used if
PEAK_MODE=1. Selects CFIR, COARSE_GAIN, or PFIR for values 0,1, or 2
respectively (the values 2 and 3 are the same). See address 28 below.
5-7
R/W
Unused
ADDRESS 27:
Input, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-2 (LSBs)
R/W
INPUT_SEL
Selects which input port to use for this channel. Values 0-3 select inputs A-D
respectively. Values 4 and 5 select differential inputs A and B. Values 6 and 7 select
the internal diagnostic source (see general address 5). Note that for differential
inputs one must also enable the differential receivers by setting the DIFF_IN bit in
register 4.
3
R/W
SEL_AB
Selects the A/B channel in the dual 12 bit plus exponent mode (INPUT_MODE=3). If
SEL_AB is 0, then the samples with their A/B flag low are used. If SEL_AB is high,
then the samples with their A/B flag high are used.
4-5
R/W
INPUT_MODE
Selects the input format: 0 is the 14 bit input mode, 1 is the 16 bit input mode, 2 is
the 12 bit plus exponent mode, and 3 is the dual 12 bit plus exponent mode.
6
R/W
Unused
7 MSB
R/W
MSB_POL
 GRAYCHIP,INC.
Invert the MSB polarity. This will convert an offset binary formatted input to 2’s
complement format.
- 40 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 28:
DATA SHEET REV 1.0
Peak Control, Suggested default = 0x1D
BIT
TYPE
NAME
DESCRIPTION
0-2 LSB
R/W
PEAK_SYNC
Synchronizes the peak counter circuitry using the mode selected in Table 7.
When the selected sync occurs, the counter transfers its contains to a read only
control register and clears the counter.
3-4
R/W
PEAK_THRESHOLD
When the input peak detect mode is selected (PEAK_MODE=0), the input word
is checked against a threshold. The threshold values are 1/4, 1/2, 3/4, or 4/4 (full
scale) for PEAK_THRESH= 0,1,2, or 3 respectively. The absolute value of the
top 11 bits is compared to the selected threshold. This is only used when the
PEAK_MODE is zero.
5
R/W
PEAK_MODE
If PEAK_MODE is zero, the input word is selected as the source to the peak
counter, otherwise one of three overflow bits (see PEAK_SELECT at address
26) is monitored by the peak counter.
6-7
R/W
Unused
ADDRESS 29:
Peak Count, Read only
BIT
TYPE
NAME
DESCRIPTION
0-7
R
PEAK_COUNT
Count of number of overflows or samples above threshold during last sync
period. The counter hard-limits at 255.
ADDRESS 30:
Fine Gain Byte 0, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-7
R/W
FINE_GAIN[0:7]
The LSBs of the fine gain control
ADDRESS 31:
Fine Gain Byte 1, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-5
R/W
FINE_GAIN[8:13]
The MSBs of the fine gain control
6-7 MSB
R/W
Unused
The fine gain control can apply a gain of FINE_GAIN/1024 providing a range of 0 to almost 16. Double buffering and
synchronization for data transfer is provided to allow the user to change the gain dynamically without causing glitches in the signal
(See GAIN_SYNC in address 22). This also allows the user to synchronously change gain in multiple channels even across
different chips.
 GRAYCHIP,INC.
- 41 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.7
DATA SHEET REV 1.0
RESAMPLER COEFFICIENT PAGES (PAGES 32-63)
These pages store the 256 resampler coefficients. Storing resampler coefficient values is similar to storing the coefficients
for the CFIR and PFIR filters. The resampler coefficients are 12 bits with the 8 LSBs written in one address, and the upper 4 bits
written as the 4 LSBs of the next address. When reading back resampler coefficients the top four bits of the second address
always read back zero.
The resampler coefficient RAM must be written in blocks of eight addresses (four coefficients). Writes to the RAM occur
when a write is done to addresses 23 or 31. Supplying coefficients in sequential order will write correctly to the RAM. If just a
portion of the resampler coefficient RAM is to be updated, then one must write in blocks of the eight addresses 16 to 23, or 24
to 31. Writing to less than eight addresses will either result in no change to the RAM or unknown changes to some coefficients.
TO LOAD A COEFFICIENT THE USER MUST WRITE IN BLOCKS OF FOUR COEFFICIENTS. ONE MUST WRITE TO
ADDRESSES 16-22 THEN ADDRESS 23 OR TO ADDRESSES 24-30 THEN ADDRESS 31.
Table 16: Resampler Coefficient Pages (Single filter mode)
Page
Address
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
16
h0
h8
h 16
h24
h32
h40
h48
h 56
h64
h72
h80
h88
h96
h 104
h112
h120
17
h0
h8
h 16
h24
h32
h40
h48
h 56
h64
h72
h80
h88
h96
h 104
h112
h120
18
h1
h9
h 17
h25
h33
h41
h49
h 57
h65
h73
h81
h89
h97
h 105
h113
h121
19
h1
h9
h 17
h25
h33
h41
h49
h 57
h65
h73
h81
h89
h97
h 105
h113
h121
20
h2
h10
h 18
h26
h34
h42
h50
h 58
h66
h74
h82
h90
h98
h 106
h114
h122
21
h2
h10
h 18
h26
h34
h42
h50
h 58
h66
h74
h82
h90
h98
h 106
h114
h122
22
h3
h11
h 19
h27
h35
h43
h51
h 59
h67
h75
h83
h91
h99
h 107
h115
h123
23
h3
h11
h 19
h27
h35
h43
h51
h 59
h67
h75
h83
h91
h99
h 107
h115
h123
24
h4
h12
h 20
h28
h36
h44
h52
h 60
h68
h76
h84
h92
h100
h 108
h116
h124
25
h4
h12
h 20
h28
h36
h44
h52
h 60
h68
h76
h84
h92
h100
h 108
h116
h124
26
h5
h13
h 21
h29
h37
h45
h53
h 61
h69
h77
h85
h93
h101
h 109
h117
h125
27
h5
h13
h 21
h29
h37
h45
h53
h 61
h69
h77
h85
h93
h101
h 109
h117
h125
28
h6
h14
h 22
h30
h38
h46
h54
h 62
h70
h78
h86
h94
h102
h 110
h118
h126
29
h6
h14
h 22
h30
h38
h46
h54
h 62
h70
h78
h86
h94
h102
h 110
h118
h126
30
h7
h15
h 23
h31
h39
h47
h55
h 63
h71
h79
h87
h95
h103
h 111
h119
h127
31
h7
h15
h 23
h31
h39
h47
h55
h 63
h71
h79
h87
h95
h103
h 111
h119
h127
Page
Address
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
16
h128
h136
h 144
h152
h160
h168
h176
h 184
h192
h200
h208
h216
h224
h 232
h240
h248
17
h128
h136
h 144
h152
h160
h168
h176
h 184
h192
h200
h208
h216
h224
h 232
h240
h248
18
h129
h137
h 145
h153
h161
h169
h177
h 185
h193
h201
h209
h217
h225
h 233
h241
h249
19
h129
h137
h 145
h153
h161
h169
h177
h 185
h193
h201
h209
h217
h225
h 233
h241
h249
20
h130
h138
h 146
h154
h162
h170
h178
h 186
h194
h202
h210
h218
h226
h 234
h242
h250
21
h130
h138
h 146
h154
h162
h170
h178
h 186
h194
h202
h210
h218
h226
h 234
h242
h250
22
h131
h139
h 147
h155
h163
h171
h179
h 187
h195
h203
h211
h219
h227
h 235
h243
h251
23
h131
h139
h 147
h155
h163
h171
h179
h 187
h195
h203
h211
h219
h227
h 235
h243
h251
24
h132
h140
h 148
h156
h164
h172
h180
h 188
h196
h204
h212
h220
h228
h 236
h244
h252
25
h132
h140
h 148
h156
h164
h172
h180
h 188
h196
h204
h212
h220
h228
h 236
h244
h252
26
h133
h141
h 149
h157
h165
h173
h181
h 189
h197
h205
h213
h221
h229
h 237
h245
h253
27
h133
h141
h 149
h157
h165
h173
h181
h 189
h197
h205
h213
h221
h229
h 237
h245
h253
28
h134
h142
h 150
h158
h166
h174
h182
h 190
h198
h206
h214
h222
h230
h 238
h246
h254
29
h134
h142
h 150
h158
h166
h174
h182
h 190
h198
h206
h214
h222
h230
h 238
h246
h254
30
h135
h143
h 151
h159
h167
h175
h183
h 191
h199
h207
h215
h223
h231
h 239
h247
h255
31
h135
h143
h 151
h159
h167
h175
h183
h 191
h199
h207
h215
h223
h231
h 239
h247
h255
Table 16 shows the coefficient register assignments when there is a single filter (NF=0). For two filters (NF=1), the two filters
are interleaved, i.e., the heven in Table 16 will contain one filter and hodd will contain the other. Four four filters (NF=3), the four
filters are interleaved, i.e., h0, h4, h8, ... is the first filter, h1, h5, ... is the second, etc.
 GRAYCHIP,INC.
- 42 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.8
DATA SHEET REV 1.0
RESAMPLER CONTROL PAGE (PAGE 64)
This page controls the resampler. The address assignments are:
Table 17: Resampler Control Registers
ADDRESS
NAME
DESCRIPTION
16
N-channels
Sets the number of channels and filters
17
N-Multiplies
Sets the number of multiplies per output
18
Filter Select
Maps channels to filter sets
19
Final Shift
Sets the final gain shift
20
Channel Map
Maps channels to outputs
21
Add To
Adds channel outputs together
22
Clock Divide
Divides the clock to the resampler
23
Ratio Map
Maps ratios to output channels
24-31
Unused
ADDRESS 16:
N-Channels Out Register, Suggested default = 0x23
BIT
TYPE
NAME
DESCRIPTION
0-1 LSB
R/W
NC
Must be set to NC=NCHAN-1, where NCHAN is the number of output channels
to be generated. A value of NC=0 means one output channel. A value of NC=1
means two output channels. Use a value of NC=3 for either three or four output
channels. A value of 2 is illegal and will produce erroneous results.
2-3
R/W
NF
Must be set to NF=NFILTER-1, where NFILTER is the number of resampler
filters. Used to partition the resampler coefficient RAM. A value of NF=0 means
one filter (normal case). A value of NF=1 means two filters. A value of NF=3
means four filters. A value of 2 is illegal.
4-6
R/W
RES_SYNC
The resampler is synchronized to this sync source (See Table 7). Resets the
delay accumulators in all channels at the same time.
7 MSB
R/W
Unused
ADDRESS 17:
N-Multiplies Register, Suggested default = 0x0E
BIT
TYPE
NAME
DESCRIPTION
0-5 LSB
R/W
NM
Must be set to NM=NMULT-1, where NMULT is the number of resampler
multiplies. The minimum legal value is NM=5, the maximum is NM=63 but
typically the maximum will be set by other constraints (see Section 3.5). In the
case of a single channel output the minimum value is NM=6.
6
R/W
NO_SYM_RES
The resampler filter is presumed to be symmetric unless this bit is set.
7 MSB
R/W
Unused
ADDRESS 18:
Filter Select Register, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-1 LSB
R/W
FILTER_SEL_0
The filter map for output channel 0. This select which of the NFILTER filters to
use for this channel. Must be less than or equal to NFILTER
2-3
R/W
FILTER_SEL_1
The filter map for output channel 1.
4-5
R/W
FILTER_SEL_2
The filter map for output channel 2.
6-7 MSB
R/W
FILTER_SEL_3
The filter map for output channel 3.
 GRAYCHIP,INC.
- 43 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 19:
DATA SHEET REV 1.0
Final Shift Register, Suggested default = 0x34
BIT
TYPE
NAME
DESCRIPTION
0-3 LSB
R/W
FINAL_SHIFT
The final shift up applied to all output channels before rounding and outputting.
Legal values are 0-15.
4-5
R/W
ROUND
Round the output to 12 (ROUND=0), 16 (ROUND=1), 20 bits (ROUND=2) or 24
bits (ROUND=3). Note, the data is output from the resampler in all cases as 24
bit words. Rounding is into the MSBs, unused LSBs of the 24 bit output words
are not cleared. The chip’s output word size is set by BITS_PER_WORD in the
output page, not by ROUND.
6
R/W
TAG_22
Two bit tag mode. Replaces the 2 LSBs of the 24 bit resampler output word with
the 2 LSBs of the tag words. Changes the 24 bit round mode (ROUND=3), if it is
used, to be a 22 bit round mode.
7 (MSB
R/W
Unused
ADDRESS 20:
Channel Map Register, Suggested default = 0xE4
BIT
TYPE
NAME
DESCRIPTION
0-1 LSB
R/W
CHAN_MAP_A
The channel map for channel A. This tells the hardware which output channel
the results of down converter channel A should be directed to.
2-3
R/W
CHAN_MAP_B
The channel map for down converter channel B.
4-5
R/W
CHAN_MAP_C
The channel map for down converter channel C.
6-7 MSB
R/W
CHAN_MAP_D
The channel map for down converter channel D.
This register maps down converter channels to output (resampler) channels. For most applications this will be a simple map
of channel A to output channel 0, channel B to output channel 1, etc. However, for multichannel modes (see Section 3.4) such
as SPLITIQ two or even four channels may be directed to the same output channel.
ADDRESS 21:
Add-To Register, Suggested default = 0x70
BIT
TYPE
NAME
DESCRIPTION
0 LSB
R/W
ADD_A_TO_B
Add down converter channel A to down converter channel B.
1
R/W
ADD_B_TO_C
Add down converter channel B to down converter channel C.
2
R/W
ADD_C_TO_D
Add down converter channel C to down converter channel D.
3
R/W
ADD_D_TO_NEXT_A
Not useful. Must be set to zero.
4-6
R/W
RATIO_SYNC
Changes to the ratio map (address 23) are synchronized to this sync source.
7 MSB
R/W
Unused
When processing complex input signals partial results are computed in adjacent channels that must be summed together to
produce a meaningful result. This control bit informs the resampler to save the data presented to it’s input and add it to the next
sample presented (if the chip is properly set up this will be from the next channel). In this manner the real and imaginary portions
of the input are rejoined prior to resampling.
ADDRESS 22:
Resampler Clock Divide Register, Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-7
R/W
RES_CLK_DIV
Resampler clock division.
In many applications only a small portion of the resampler computational throughput is required. Power can be reduced by
dividing the clock driving the resampler. The resampler clock rate is 2*FCK/(1+RES_CLK_DIV). Caution must be used to avoid
 GRAYCHIP,INC.
- 44 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
dividing the clock so far that there is not enough clock cycles to complete the computations (see Section 3.5). It is recommended
that an application first be brought up without resampler clock division.
ADDRESS 23:
Ratio Map Register Suggested default = 0x00
BIT
TYPE
NAME
DESCRIPTION
0-1 LSB
R/W
RATIO_MAP_0
The ratio map for channel 0. This tells the hardware which resampler ratio
should be use for output channel 0.
2-3
R/W
RATIO_MAP_1
The ratio map for output channel 1.
4-5
R/W
RATIO_MAP_2
The ratio map for output channel 2.
6-7 MSB
R/W
RATIO_MAP_3
The ratio map for output channel 3.
The default ratio maps select ratio 0 for output 0, ratio 1 for output 1, ratio 2 for output 2, and ratio 3 for output 3. The ratio
maps can also be used to synchronously switch between resampling ratios. This allows the chip’s resampler to be used in timing
loops where the ratio must toggle between several values which have been programmed into the chip.
5.9
RESAMPLER RATIO PAGE (PAGE 65)
This page stores four resampler ratios to be used by the resampler channels. Each ratio is a 32 bit ratio of the input sample
rate to the output sample rate with an implicit decimal point six bits down from the top. The total range for the ratio is then 0 to
63. The hardware limits the decimation to be less than 32 (hence the MSB of the 32 bit word should always be zero).
Table 18: Resampler Ratio Page
5.10
ADDRESS
NAME
ADDRESS
NAME
16
RATIO_0 (LSBs)
24
RATIO_2 (LSBs)
17
RATIO_0
25
RATIO_2
18
RATIO_0
26
RATIO_2
19
RATIO_0 (MSBs)
27
RATIO_2 (MSBs)
20
RATIO_1 (LSBs)
28
RATIO_3 (LSBs)
21
RATIO_1
29
RATIO_3
22
RATIO_1
30
RATIO_3
23
RATIO_1 (MSBs)
31
RATIO_3 (MSBs)
CHANNEL OUTPUT PAGES (PAGE 96 & 97)
Addresses 16 through 31 on these pages are used to read output values. The outputs are 24 bit two’s complement numbers
which are read as three 8 bit bytes. Reading address 16 will output 0 on C[0:7]. Reading address 17 will output the least
significant byte. Address 18 provides the middle byte. Address 19 provides the most significant byte. If microprocessor mode is
enabled, and pin P0-23 are enabled (by setting EN_P0-3, EN_PAR) the full 24 bit word is simultaneously output P[0:23].These
are all read only registers.
See Tables 4, 5 and 6 for the data order in these pages.
 GRAYCHIP,INC.
- 45 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
5.11
DATA SHEET REV 1.0
OUTPUT CONTROL PAGE (PAGE 98)
This page controls the output. The following table summarizes the registers:
Table 19: Output Control Registers
ADDRESS
NAME
DESCRIPTION
16
Tristate Controls
Enables serial and parallel ports and controls.
17
Output Format
Sync output. Invert SCK, SFS, and RDY. Set RDY width.
Enable tags.
18
Output Mode
Select serial, nibble, parallel, or uP output. Select real or
complex output. Enable nibble reverse.
19
Output Frame Control
Select output frame length. Control SFS behavior.
20
Output Word Sizes
Select output word size, output block size, and word per
frame.
21
Output Clock Control
Set clock divider for serial clock. Set number of active
serial outputs.
22
Serial Mux Control
Route serial streams to serial ports.
23
Output Tag A
Tags for outputs AI and AQ
24
Output Tag B
Tags for outputs BI and BQ
25
Output Tag C
Tags for outputs CI and CQ
26
Output Tag D
Tags for outputs DI and DQ
27
Mask Revision
Reads back the chip’s mask revision number
28
Miscellaneous
Output enable for SO, four frame strobe mode control
ADDRESS 16:
Tristate Controls, Cleared on power up, Suggested default = (see Table 3)
BIT
TYPE
NAME
DESCRIPTION
0 (LSB)
R/W
EN_SCK
Enable SCK output. Used in serial, nibble, LINK, and parallel modes.
1
R/W
EN_RDY
Enable RDY output. Used for parallel and uP modes. RDY should be enabled for
the master chip in serial or nibble TDM modes. RDY should not be enabled for
slave chips, or for the LINK mode.
2
R/W
EN_SFS
Enable SFS output. Used in serial, nibble, and parallel modes. Can be used as
an interrupt in LINK mode.
3
R/W
EN_P0
Enable P0. Required for serial mode when outputting on P0, LINK or Nibble
modes, wide word uP, and parallel modes.
4
R/W
EN_P1
Enable P1. Required for serial mode when outputting on P1, LINK or Nibble
modes, wide word uP, and parallel modes.
5
R/W
EN_P2
Enable P2. Required for serial mode when outputting on P2, LINK or Nibble
modes, wide word uP, and parallel modes.
6
R/W
EN_P3
Enable P3. Required for serial mode when outputting on P3, LINK or Nibble
modes, wide word uP, and parallel modes.
7 (MSB)
R/W
EN_PAR
Enable P4-P23. Used for parallel output and wideword uP modes.
All outputs except SO and the control port (C0-7) power up into a high impedance, tristate, mode until they are enabled. The
control port (C0-7) is tristate until both CE and RD are low and WR is high.
 GRAYCHIP,INC.
- 46 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 17:
DATA SHEET REV 1.0
Output Format, Suggested default = 0x40
BIT
TYPE
NAME
DESCRIPTION
0 (LSB)
R/W
INV_SCK
Invert SCK output. The serial, nibble, link and parallel outputs normally change
on the rising edge of SCK. If INV_SCK is set high the clock is inverted so data
changes on the falling edge.
1
R/W
INV_RDY
Invert RDY output. RDY is normally active high. This control bit is used for uP
and parallel modes. It has no effect in serial, LINK, or nibble modes.
2
R/W
INV_SFS
Invert SFS output. SFS is normally active high.
3
R/W
TAG_EN
Enable tags. The four bit tags replace the four LSBs of the data when TAG_EN
is set. The BITS_PER_WORD (address 20) control determines the tag location.
4
R/W
RDY_WIDTH
The RDY pulse is 4 (RDY_WIDTH=0) or 16 (RDY_WIDTH=1) CK cycles. Valid
for the uP mode only, must be set to 0 for all other modes.
5-7 (MSB)
R/W
OUT_BLK_SYNC
Output circuit sync source. See table 7 for the sync modes.
The OUT_BLK_SYNC is used to synchronize the output timing among multiple
GC4016 chips. The OUT_BLK_SYNC should only be made active during
initialization. Use during operation may cause unknown output transients.
ADDRESS 18:
Output Mode Cleared on powerup., Suggested default = (see Table 3)
BIT
TYPE
NAME
DESCRIPTION
0 (LSB)
R/W
LINK
Enable LINK protocol. This mode supports Analog Devices LINK protocol. In
this mode the RDY pin serves as LINK Acknowledge and is an input to the
GC4016 (EN_RDY must be low). NIBBLE must also be set high (eight bit LINK
mode is not supported) and OUTPUT_MODE must be set to 2. The PARALLEL
control bit must be low.
1
R/W
NIBBLE
Enable nibble mode. This mode is similar to serial except the four pins P0-3
output one nibble at a time. TDM of several chips is supported with one acting
as the master (EN_RDY=MASTER=1) and the others as the slave
(EN_RDY=MASTER=0).
2
R/W
PARALLEL
Enable parallel mode.
3
R/W
MASTER
Normally set high. Is set low by slave chips in TDM serial or nibble modes. The
RDY pin is an output when MASTER=1, and is an input when MASTER=0.
4
R/W
REAL_ONLY
Normally set low. Is set high when outputting real, instead of complex, data.
Normally used with the complex to real modes of the PFIR (see Section 3.3.7).
5-6
R/W
OUTPUT_MODE
The output mode selection is:
0 for microprocessor mode,
1 for serial mode,
2 for nibble or LINK modes, or
3 for parallel mode.
Note that only one mode is supported at a time.
7 (MSB)
R/W
REVERSE_IQ
Used when OUTPUT_ORDER=0 to swap I and Q. This is useful for link or
nibble mode outputs when packing two 16 bit words into a 32 bit transfer. See
Section 3.8.5.
 GRAYCHIP,INC.
- 47 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 19:
DATA SHEET REV 1.0
Output Frame Control, Suggested default = (see Table 3)
BIT
TYPE
NAME
DESCRIPTION
0-5 (LSB)
R/W
FRAME_LENGTH
Used in the serial and nibble modes to set the frame length, and is not used in
the microprocessor, link or parallel modes.
Serial, nibble modes:
The output frame length is (FRAME_LENGTH+1) in words (not complex
pairs). The frame length must be equal or greater than the number of
output words on the serial port. Values larger than the number of output
words can be useful to smooth the data output rate by slowing down the
output frame rate. The output frame rate must be at least as fast as the
average output data rate.
In the 16 bit nibble mode (BITS_PER_WORD=0), FRAME_LENGTH must
be greater than or equal to one (a frame length of at least two words).
If this is the master chip in a multi-chip TDM mode (FRAME_LENGTH+1)
sets the TDM frame length. The master chip will occupy the first
(WORDS_PER_FRAME+1) time slots of the frame.
If this is a slave chip, then (FRAME_LENGTH+1) is the delay, in number of
words, from the start of frame before outputting its data.
Microprocessor, link or parallel modes:
Unused, set to 0
6-7 (MSB)
R/W
SFS_MODE
Used in the serial, nibble and parallel modes to set the frame strobe modes. Not
used in the microprocessor or link modes.
Serial mode: (See Figure 15)
In the serial mode SFS is active for one SCK cycle ahead of the first bit.
If SFS_MODE=0 (or = 2), then SFS is active for one SCK cycle at the start
of the frame.
If SFS_MODE=1, then SFS is active once for each I word.
If SFC=3 then SFS is active for each word (I or Q).
Link mode:
The SFS behavior in the nibble mode is similar except the frame strobe
occurs concurrent with the first nibble rather than 1 SCK cycle early.
Parallel mode: (See Figure 16)
In the parallel mode, SFS and RDY are used as flags to indicate valid
output data.
If SFS_MODE=0 or 1, then SFS is active for one SCK cycle when the I
sample is valid and RDY is active for one SCK cycle when Q is valid.
If SFS_MODE=2 or 3, then SFS is active for one SCK cycle when either I
or Q are valid and RDY is a start of frame signal which is active for when
the first I sample of an output block is valid (see BLOCK_SIZE address 20
and Figure 16). Note that when SFS_MODE=2 or 3 the RDY flag goes high
after the last word in a frame and goes low after the first word in the next
frame.
Microprocessor or link modes:
Unused, set to 0
 GRAYCHIP,INC.
- 48 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 20:
DATA SHEET REV 1.0
Output Word Sizes, Suggested default = (see Table 3)
BIT
TYPE
NAME
DESCRIPTION
0-2 (LSB)
R/W
WORDS_PER_FRAME
Serial, nibble and parallel modes:
The number of output words (not complex pairs) in a frame is
(WORDS_PER_FRAME+1). This value should be less than or equal to the
frame length. It should also be less than or equal to the number of output
channels.
Serial or nibble TDM modes:
(WORDS_PER_FRAME+1) is the number of words THIS chip inserts into
the TDM stream.
Link mode:
In the link mode only 32 bit transfers are understood. A single 32 bit word is
transferred by setting WORDS_PER_FRAME to 0 and BITS_PER_WORD
to 1. Two 16 bit words packed together in a single 32 bit transfer is done by
setting WORDS_PER_FRAME to 1 and BITS_PER_WORD to 0.
Microprocessor mode:
Unused, set to 0.
3-5
R/W
BITS_PER_WORD
Sets the number of bits per word in the serial, nibble and link modes and
determines the location of tag bits in all modes.
Serial mode:
For serial outputs the number of bits per serial word is set as
4*(BITS_PER_WORD+1), where BITS_PER_WORD ranges from 2 to 7.
Values 2-5 create 12-24 bit serial outputs with tags (if enabled) in the least
significant nibble. Values 6 and 7 create 28 and 32 bit outputs respectively
with the last one or two nibbles zeros and tags in the sixth nibble.
Nibble and link modes:
For nibble and link modes the number of bits per word are 16 for
BITS_PER_WORD=0 and 32 for BITS_PER_WORD=1. Tag bits replace
the least significant nibble when BITS_PER_WORD=0, and replace the
sixth nibble when BITS_PER_WORD=1. The least significant two nibbles
are always zero when BITS_PER_WORD=1.
Parallel or microprocessor modes:
For parallel or uP outputs BITS_PER_WORD only affects the placement of
the tag bits. The least significant four bits of the 4*(BITS_PER_WORD+1)
bits are replaced. BITS_PER_WORD ranges from 2 to 5 in these modes.
Note that rounding is independent and is controlled by the ROUND control in
address 19 of the resampler control page.
6-7 (MSB)
R/W
BLOCK_SIZE
Outputs are transferred from the resampler to the output in blocks of size
(BLOCK_SIZE+1) complex samples (two words per sample). Values 0,1, and 3
are valid. Value 2 is mapped to be the same as 3.
Synchronous serial, nibble and parallel modes:
For synchronous serial, nibble and parallel outputs the block size will
typically match the number of output channels.
Link mode and asynchronous serial, nibble and parallel modes:
For link and asynchronous outputs the block size should be minimum
(BLOCK_SIZE=0).
Microprocessor mode:
For the microprocessor modes the BLOCK_SIZE should be 3 which will
minimize the interrupt rate to one interrupt per 4 samples (real or complex).
 GRAYCHIP,INC.
- 49 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 21:
DATA SHEET REV 1.0
Output Clock Control, Suggested default = (see Table 3)
BIT
TYPE
NAME
DESCRIPTION
0-3 (LSB)
R/W
SCK_RATE
Serial clock rate is SCK = CK/(1+SCK_RATE). SCK_RATE can be 0 to 15. If
SCK_RATE=0, then the serial clock rate will be equal to CK. SCK is also the
output clock for nibble and parallel modes.
4-5
R/W
NSERIAL
The number of serial pins used to output the data is NSERIAL+1. Values 0,1,
and 3 are valid (value 2 is mapped to be the same as value 3). A value of 0
means that all outputs will be multiplexed onto one serial stream. See Figure
15(c).
Must be 0 for nibble, and link modes.
Unused for uP, and parallel modes.
6-7 (MSB)
R/W
OUTPUT_ORDER
OUTPUT_ORDER is normally set to 0 which will cause the data to be output in
the same order as it is computed by the resampler. If the channels are
synchronous, then the order will be IA,QA,IB,QB,IC,QC,ID,QD.
OUTPUT_ORDER must be set to 0 for asynchronous data.
OUTPUT_ORDER equal to 1 or 2 is only valid for synchronous channel data.
These modes allow channels B, C and D to be powered up and down without
disturbing the channel order. Channel A can not be powered down.
OUTPUT_ORDER=1 will cause the output order to be
IA,IB,QA,QB,IC,ID,QC,QD. This is appropriate for complex synchronous serial
data output on two streams.
OUTPUT_ORDER=2 (or 3) will cause the output order to be
IA,IB,IC,ID,QA,QB,QC,QD. This is appropriate for synchronous serial data
output on four 4 serial streams.
See Figure 15(c).
ADDRESS 22:
Serial Mux Control, Suggested default = 0xE4
BIT
TYPE
NAME
DESCRIPTION
0-1 (LSB)
R/W
SMUX_0
Serial stream selection for serial pin P0. A four to one multiplexer allows serial
streams SOUTA (SMUX_0=0), SOUTB (SMUX_0=1), SOUTC (SMUX_0=2)
and SOUTD (SMUX_0=3) to be routed to serial pin P0. See Figure 15 for the
definition of serial streams SOUTA, SOUTB, SOUTC and SOUTD.
If SMUX_0 is set to 1 in the LINK or NIBBLE output modes then the bit order
within the nibble is reversed (nibble LSB will be on P3 rather than P0). This
allows backwards compatibility with the GC4014 in LINK mode.
2-3
R/W
SMUX_1
Serial stream selection for serial pin P1 as above. Note that the same serial
stream may be routed to more than one serial pin if desired.
4-5
R/W
SMUX_2
Serial stream selection for serial pin P2.
6-7 (MSB)
R/W
SMUX_3
Serial stream selection for serial pin P3.
 GRAYCHIP,INC.
- 50 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
ADDRESS 23:
DATA SHEET REV 1.0
Output Tag A, Suggested default = 0x10
BIT
TYPE
NAME
DESCRIPTION
0-3 (LSB)
R/W
TAG_AI
Four bit tag for serial stream A, I word.
4-7 (MSB)
R/W
TAG_AQ
Four bit tag for serial stream A, Q word.
ADDRESS 24:
Output Tag B, Suggested default = 0x32
BIT
TYPE
NAME
DESCRIPTION
0-3 (LSB)
R/W
TAG_BI
Four bit tag for serial stream B, I word.
4-7 (MSB)
R/W
TAG_BQ
Four bit tag for serial stream B, Q word.
ADDRESS 25:
BIT
Output Tag C, Suggested default = 0x54
TYPE
NAME
DESCRIPTION
0-3 (LSB)
R/W
TAG_CI
Four bit tag for serial stream C, I word.
4-7 (MSB)
R/W
TAG_CQ
Four bit tag for serial stream C, Q word.
ADDRESS 26
BIT
Output Tag D, Suggested default = 0x76
TYPE
NAME
DESCRIPTION
0-3 (LSB)
R/W
TAG_DI
Four bit tag for serial stream D, I word.
4-7 (MSB)
R/W
TAG_DQ
Four bit tag for serial stream D, Q word.
Note: in the real output mode (REAL_ONLY=1), only the Q tags will be used.
ADDRESS 27:
Mask Revision, Read Only
BIT
TYPE
NAME
DESCRIPTION
0-7
R
REVISION
Mask revision number.
This address allows the user to read the current mask revision code from software. See Section 3.16 for details.
ADDRESS 28
Miscellaneous Controls, Suggested default = 0x02, Cleared by power up.
BIT
TYPE
NAME
DESCRIPTION
0 (LSB)
R/W
EN_4_FS
Enables the four frame strobe mode. The four frame strobe mode is used with
asynchronous serial or nibble data to identify words from each channel. TAG_22
in address 19 of the resampler control page must also be set. The two LSBs of
the tag words are used to generate the frame strobes. The output timing of the
frame strobes are the same as for SFS. If the Tag is 0, then a copy of SFS is
output as FSA, if the tag is 1, then a copy of SFS is output on FSB, if the tag is 2
then a copy of SFS is output on FSC, and if the tag is 3, then a copy of SFS is
output on FSD.
1
R/W
EN_SO
The output enable control for SO
3-7 (MSB)
R/W
unused
 GRAYCHIP,INC.
- 51 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
THIS PAGE IS INTENTIONALLY BLANK
 GRAYCHIP,INC.
- 52 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
6.0
SPECIFICATIONS
6.1
ABSOLUTE MAXIMUM RATINGS
DATA SHEET REV 1.0
Table 20: Absolute Maximum Ratings
CAUTION: 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.
PARAMETER
SYMBOL
MIN
MAX
UNITS
Pad Ring Supply Voltage
VPAD
-0.3
4.1
V
Core Supply Voltage
V CORE
-0.3
3.0
V
Input voltage (undershoot and overshoot)
V IN
-0.5
VPAD+0.5
V
Storage Temperature
TSTG
-65
150
Junction Temperature under operation
TJ
°C
°C
°C
125
Lead Soldering Temperature (10 seconds)
NOTES
300
1
ESD Classification
Class 3A Human Body Model (4 kV) (JESD22-A114-B)
Class 4 Charged Device Model (1 kV) (JESD22-C101-A)
Moisture Sensitivity
Class 2
1. The circuit is designed for junction temperatures up to 125C. Sustained operation above 125C junction temperature will reduce long term
reliability.
6.2
RECOMMENDED OPERATING CONDITIONS
Table 21: Recommended Operating Conditions
PARAMETER
SYMBOL
MIN
MAX
UNITS
NOTES
Pad Ring Supply Voltage
VPAD
3.0
3.6
V
1
Core Supply Voltage
V CORE
2.3
2.7
V
1
Temperature Ambient, no air flow
TA
-40
+85
Junction Temperature
TJ
°C
°C
100
1
2
1. DC and AC specifications in Tables 23 and 24 are production tested over these ranges.
2. Thermal management may be required for full rate operation, See Table 22 below and Section 7.4.
6.3
THERMAL CHARACTERISTICS
Table 22: Thermal Data
160 PBGA
THERMAL CONDUCTIVITY
SYMBOL
UNITS
0.5 Watt
Theta Junction to Ambient
Theta Junction to Case
θJA
θJC
1 Watt
TBD: Estimated at 32
TBD
TBD
°C/W
°C/W
Note: Air flow will reduce θja and is highly recommended.
 GRAYCHIP,INC.
- 53 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
6.4
DATA SHEET REV 1.0
POWER CONSUMPTION
The maximum power consumption is a function of the operating mode of the chip. The following equation estimates the
typical power supply current for the chip. Chip to chip variation is typically +/- 5%. Table 24 provides maximum current in a
maximum configuration used in production test.
o u t
I PAD (TYP) = ( V PAD )  F
------- ( N o u t ) ( C o u t + 2pF )
 4 
V CORE FCK 
225 23

I CORE (TYP) =  -------------  ---------- + ------ mA
 2.5   80M 10 + A  31 + -------N R
Where A is the number of active channels (0 to 4), N is the CIC decimation ratio, R is the resampler clock division, Fout is the
output SCK rate, Nout is the number of active output data pins
, and Cout is the average capacitive load on each data pin. The equation assumes random data transition density of 1 rising
edge per four SCK cycles.
6.5
DC CHARACTERISTICS
Table 23: DC Operating Conditions (-40 to 85 °C case unless noted)
VPAD = 3.0 to 3.6V
PARAMETER
MIN
Voltage input low
VIL
Voltage input high
VIH
Voltage output low (IOL = 2mA)
VOL
Voltage output high (IOH = -2mA)
VOH
Leakage current (V IN = 0V or VPAD)
Inputs or Outputs in tristate condition
UNITS
TEST
LEVEL
V
IV
V
IV
V
IV
V
IV
1
uA
IV
35
uA
IV
2
mA
IV
SYMBOL
MAX
0.8
2.0
0.5
2.4
IIN
Pullup current (VIN = 0V) (TDI, TMS, TCK)
| I PU |
5
Quiescent supply current, I CORE or IPAD
(V IN=0 or V IN=VPAD, Address 0 = F0, DIFF_IN=0)
ICCQ
Data input capacitance (All inputs except CK)
CIN
4 (typical)
pF
I
Clock input capacitance (CK input)
CCK
13 (typical)
pF
I
Notes:
Currents are measured at nominal voltages, high temperature (85C).
Voltages are measured at low speed. Output voltages are measured with the indicated current load
Test Levels:
I. Controlled by design and process and not directly tested or recommended practice.
II. Verified on initial part evaluation.
III. 100% tested at room temperature, sample tested at hot and cold.
IV. 100% tested at hot, sample tested cold.
V. 100% tested at hot and cold.
 GRAYCHIP,INC.
- 54 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
6.6
DATA SHEET REV 1.0
AC CHARACTERISTICS
Table 24: AC Characteristics (-40 TO +85oC Case, across recommended voltage range, unless noted)
PARAMETER
Clock Frequency
Clock low or high period
SYMBOL
MIN
MAX
UNITS
TEST
LEVEL
FCK
Note 1
100
MHz
IV
tCKL/H
3
ns
IV
70
%
II
2
ns
I
Clock Duty Cycle (t CKH as a percentage of the clock period)
Clock rise and fall times (VIL to VIH )
tRF
Input setup before CK goes high
(AIN, BIN, CIN, DIN, or SI) Single ended
tSU
2
ns
IV
Input hold time after CK goes high
tHD
0.8
ns
IV
Data output delay from rising edge of CK.
(P[0:23], SFS, SCK, RDY, or SO)
tDLY
1
5
ns
IV
Data output skew from rising edge of SCK.
(P[0:23], SFS, or RDY)
tSKEW
-2.5
2.5
ns
IV
JTAG Clock Frequency
FJCK
40
MHz
IV
JTAG Clock low period (Below VIL)
tJCKL
10
ns
IV
JTAG Clock high period (Above V IH)
tJCKH
10
ns
IV
JTAG Input (TDI or TMS) setup before TCK goes high
tJSU
1
ns
IV
JTAG Input (TDI or TMS) hold time after TCK goes high
tHD
10
ns
IV
JTAG output (TDO) delay from rising edge of TCK.
tDLY
ns
IV
Control setup during reads or writes.See Figures 2 and 3.
tCSU
2
ns
IV
tEWCSU
4
ns
IV
tCHD
1
ns
IV
Control strobe (CS and WR low) pulse width
(Write operation). See Figures 2 and 3.
tCSPW
20
ns
IV
Control output delay CS and RD low and A stable to C
(Read Operation). See Figures 2 and 3.
tCDLY
12
ns
IV
Control recovery time between reads or writes. See Figures 2 and 3.
tREC
20
ns
I
Core dynamic supply current nominal voltages, 100 MHz, 4 channels
active, resampler at full speed, high temperature.
ICDYN
390
mA
IV
Control data setup during writes (edge mode). See Figure 3.
Control hold during writes. See Figures 2 and 3.
10
100
Notes:
1. The minimum clock rate must satisfy FCK/(4N) > 10KHz, where N is the CIC decimation ratio.
2. Timing between signals is measured from mid-voltage (VPAD/2) to mid-voltage. Output loading is a 50 Ohm transmission line.
Test Levels:
I. Controlled by design and process and not directly tested or recommended practice.
II. Verified on initial part evaluation.
III. 100% tested at room temperature, sample tested at hot and cold.
IV. 100% tested at hot, sample tested cold.
V. 100% tested at hot and cold.
 GRAYCHIP,INC.
- 55 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
7.0
APPLICATION NOTES
7.1
POWER AND GROUND CONNECTIONS
The GC4016 chips are very high performance chips which require solid power and ground connections to avoid noise on
the VCORE, V PAD and GND pins. If possible the GC4016 chip should be mounted on a circuit board with dedicated power and
ground planes and with at least two decoupling capacitors (0.01µf) adjacent to each side of the chip, one for VCORE and one for
VPAD.
IMPORTANT
THE GC4016 CHIP MAY NOT OPERATE PROPERLY IF THESE POWER AND GROUND GUIDELINES ARE VIOLATED.
7.2
STATIC SENSITIVE DEVICE
The GC4016 chips are fabricated in a high performance CMOS process which is sensitive to the high voltage transients
caused by static electricity. These parts can be permanently damaged by static electricity and should only be handled in static
free environments. See Table 20.
7.3
MOISTURE SENSITIVE PACKAGE
The GC4016 come in level 2 moisture sensitive packages. Dry pack storage is required prior to assembly. If parts are stored
out of dry pack for more than 72 hours, then parts must be baked for 24 hours at 100°C prior to assembly.
7.4
THERMAL MANAGEMENT
The parameters in Section 6.0 are tested at a case temperature of 100°C. In any case, the junction temperature must be
kept below 125
°C for reliable operation. To determine the junction temperature, the user should calculate the chip’s power
dissipation using the equation for supply current in Section 6.4 and then use the package’s thermal conductivity shown in Section
6.3. The junction temperature is calculated by adding the operating ambient temperature (or case temperature) to the product of
the power consumption times the thermal conductivity.
For example, the GC4016 chip operating in the GSM mode described in Section 7.9, consumes 0.3 Watts of power. The
junction to ambient rise of the PBGA160 package is 32 degrees (TBD) per Watt. This represents a rise of 10 degrees over
ambient. This means that under these conditions the ambient temperature has to be less than 90°C to keep the junction
temperature below 100°C. Air flow will decrease the thermal resistance by 10% to 40%, allowing ambient temperatures up to
96°C. Increasing the decimation ratio (N) or decreasing the number of active channels (A) will also allow a higher ambient
temperature.
 GRAYCHIP,INC.
- 56 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.5
DATA SHEET REV 1.0
EXAMPLE CFIR FILTER SETS
The CFIR filter must be flat over the output passband of interest, and have a stopband which rejects out of band signals
which will fall into the passband of interest. The spectral requirements of this filter are shown in Figure 10 in Section 3.3.5. The
CFIR filter must also compensate for the passband roll off of the CIC filter. The following filters compensate for the CIC roll off
and provide passband bandwidths which are between 17% and 150% of the down converter’s output sample rate (the PFIR
output rate before resampling).
The filter taps can be copied from this table, or requested by e-mail from [email protected], or downloaded from
the web at www.graychip.com. The filter taps are available in two file formats. The first, <filter_name>.taps, is an ascii listing of
the taps. The second, <filter_name>.cmd, is a command file which could be used to write the filter taps into a GC4016 chip. The
commands are formatted as “write <address> <data_byte>,” where <address> and <data_byte> are unsigned hex integers.
The recommended default filter is cfir_80 which is flat out to 80% of the channel output’s usable bandwidth. The narrower
filters have less ripple and better out of band rejection. The wider filters (100% and 150%) are designed for the single wideband
output per chip mode where all four channels are used as one wideband channel. These modes have more ripple and less out
of band rejection than the narrowband filters.
Table 25: Example CFIR Filters
Name
Usable
Output
Bandwidth
Passband
Passband
Stopband Stopband CFIR Gain
Ripple
Cutoff
Start
Rejection (CFIR_SUM)
(pk to pk)
Applications
Taps (first 11)
cfir_17
17%
0.085
0.01dB
1.5
110dB
2.0413
(133781)
8x oversampled outputs.
5 39 -166 -1301 -3009 -2848
1280 9009 18952 28546 32767
cfir_34
34%
0.17
0.01dB
1.5
110dB
1.5429
(101113)
4x oversampled outputs, -24 74 494 548 -977 -3416
2X GSM or EDGE outputs -3672 1525 13074 26547 32767
cfir_68
68%
0.34
0.02dB
1.55
100dB
1.2665
(83001)
2x oversampled outputs, 12 -93 -62 804 1283 -1273
1X GSM or EDGE outputs -5197 -3512 8332 24823 32767
cfir_80
(default)
80%
0.4
0.025dB
1.5
100dB
0.9899
(64877)
1.5x oversampled
outputs, Default filter for
most applications
cfir_100
100%
0.5
0.08dB
1.5
100dB
0.9109
(59695)
Single wideband DDC
-3 40 -75 -437 291 1951 -489
mode (See Section 3.4.2) -6365 -1185 19736 32767
cfir_150
150%
0.75
0.6dB
1.25
60dB
0.9154
(59993)
Single wideband DDC
47 -102 -1184 -1274 1487 3243
mode (See Section 3.4.2) -1798 -7760 15 20939 32767
5 -8 -172 -192 806 1493 -1889
-6182 1085 21109 32767
Key:
The Usable Output Bandwidth is the passband bandwidth expressed as a percentage of the DDC channel output sample rate.
The Passband Cutoff frequency is relative to the DDC channel output sample rate.
The Stopband Start frequency is relative to the DDC channel output sample rate.
The Taps are the first 11 taps of the 21 tap filter. The final 10 taps are a mirror image of the first 10 taps. The center taps are scaled to the full
scale positive value of 32767.
The Gain is CFIR_SUM/65536, the value in parenthesis is the CFIR_SUM
 GRAYCHIP,INC.
- 57 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.6
DATA SHEET REV 1.0
EXAMPLE PFIR FILTER SETS
The spectral requirements of the PFIR filter are shown in Figure 11 in Section 3.3.6. The filters in Table 26 provide passband
bandwidths which are between 17% and 150% of the down converter channel output sample rate (before resampling).
The filter taps can be copied from this table, or requested by e-mail from [email protected], or downloaded from
the web at www.graychip.com. The filter taps are available in two file formats. The first, <filter_name>.taps, is an ascii listing of
the taps. The second, <filter_name>.cmd, is a command file which could be used to write the filter taps into a GC4016 chip. The
commands are formatted as “write <address> <data_byte>”, where <address> and <data_byte> are unsigned hex integers.
Table 26: Default PFIR Filters
Name
Usable
Output
Bandwidth
Passband
Stopband
Passband
Stopband
Ripple
Rejection PFIR Gain Applications
Cutoff
Start
(pk to pk)
near/far
pfir_17
17%
0.085
0.01dB
0.125
97dB/
120dB
3.1262
(204881)
8x oversampled
outputs.
6 17 33 49 54 33 -28 -132 -263 -380 -424
-329 -49 408 964 1461 1690 1445 597
-826 -2589 -4252 -5228 -4908 -2819 1222
7006 13910 20982 27118 31289 32767
pfir_34
34%
0.17
0.02dB
0.3
83dB/
108dB
2.2370
(146605)
4x oversampled
outputs,
2X GSM or EDGE
outputs
14 30 41 27 -29 -118 -200 -212 -95 150
435 598 475 5 -680 -1256 -1330 -653 669
2112 2880 2269 101 -2996 -5632 -6103
-3091 3666 13042 22747 30053 32767
pfir_68
68%
0.34
0.015dB
0.5
90dB/
98dB
1.2140
(79563)
2x oversampled
outputs, 1X GSM
or EDGE outputs
2 1 -11 -23 -2 45 43 -48 -117 -8 191 155
-189 -375 26 579 358 -601 -918 248 1469
618 -1680 -1995 1104 3690 845 -5354
-5506 6574 24277 32767
pfir_80
80%
0.4
0.15dB
0.5
78dB/
93dB
1.1494
(75329)
1.5x oversampled
outputs, Default
filter for most
applications
31 136 208 107 -123 -181 69 277 46 -353
-235 358 483 -246 -750 -24 967 470 -1046
-1081 884 1817 -360 -2608 -689 3365
2594 -3992 -6527 4407 23277 32767
pfir_100
100%
0.5
0.2dB
0.6
75dB/
84dB
0.9189
(60221)
Single wideband
DDC mode (See
Section 3.4.2)
-30 -141 -180 2 151 -34 -186 87 232 -171
-273 293 292 -458 -270 668 181 -916 4
1192 -325 -1479 837 1758 -1647 -2005
3015 2201 -5914 -2326 19169 32767
pfir_150
150%
0.75
0.75dB
0.82
50dB/
65dB
0.6320
(41417)
Single wideband
DDC mode (See
Section 3.4.2)
-32 -234 98 -24 -74 165 -199 138 13 -198
329 -323 149 145 -437 577 -456 73 445
-867 955 -576 -213 1138 -1778 1706 -659
-1331 3909 -6474 8360 32767
Taps (first 32)
Key:
The Usable Output Bandwidth is the passband bandwidth expressed as a percentage of the DDC channel output sample rate.
The Passband Cutoff frequency is relative to the DDC channel output sample rate.
The Stopband Start frequency is relative to the DDC channel output sample rate.
The Taps are the first 32 taps of the 63 tap filter. The final 31 taps are a mirror image of the first 31 taps. The center taps are scaled to the full
scale positive value of 32767.
The Gain is PFIR_SUM/65536, the value in parenthesis is the PFIR_SUM
The recommended default filter is PFIR_80 which is flat out to 80% of the channel output’s usable bandwidth. The narrower
filters have less ripple and better out of band rejection. The wider filters (100% and 150%) are designed for the single wideband
output per chip mode where all four channels are used as one wideband channel. These modes have more ripple and less out
of band rejection than the narrowband filters.
The PFIR can also be used to apply receive filters for digital modulation formats such as QPSK, OQPSK or QAM. The most
common receive filter is the root-raised cosine (RRC) filter. A program (written in C) designed to generate RRC pfir filters is
available from graychip by e-mail or from the web. This program asks for the input sample rate, the baud rate and the desired
roll off (alpha) factor. The output is a 63 tap PFIR filter tap file and a GC4016 command file.
 GRAYCHIP,INC.
- 58 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
cfir_17 with pfir_17
cfir_34 with pfir_34
cfir_68 with pfir_68
cfir_80 with pfir_80
cfir_100 with pfir_100
cfir_150 with pfir_150
Figure 19. Overall Spectral Responses of Example Filters
 GRAYCHIP,INC.
- 59 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.7
DATA SHEET REV 1.0
EXAMPLE RESAMPLER CONFIGURATIONS
This Section describes some commonly used resampler configurations and the filter coefficient sets used for them.
7.7.1 Bypass Mode
The resampler is bypassed by using a configuration which has h0 set to 1024, all other taps set to zero, NMULT set to 7,
NO_SYM_RES set to 1, FINAL_SHIFT set to 5, and RATIO set to 226 (0x04000000). Note that the NDELAY term in the
RES_GAIN equation does not apply in this case and should be set to unity in the gain equation. The resampler control register
setting for this mode are shown in Table 27.
Table 27: Resampler Bypass Mode
Address
16
17
18
Page 32
00
04
00
Pages 33-63
00
19
20
21
22
23
24
25
Page 64
23
46
00
35
E4
70
00
00
unused
Page 65
00
00
00
04
00
00
00
04
00
00
26
27
28
29
30
31
00
04
00
00
00
04
7.7.2 Resampling Modes
Several resampler configurations are summarized in Table 28. These configurations use the default values for the resampler
controls except for NMULT in address 17, which must be set to the value in Table 28. The defaults (hex) are 23, NM, 00, 34, E4,
00, 00 and 00, respectfully for addresses 16 through 23 of page 64. NM is NMULT-1.
Table 28: Resampler Modes
Configuration
Name
NMULT
%BW
PB
Ripple
(dB)
Image
Rejection
(dB)
RES_SUM
Application
Configurations with NDELAY = 64,
These offer the best resampling phase jitter performance (-48dB inband noise), but the % input bandwidth must be 60% or less
res_6x64_60
6
res_7x64_60
7
res_8x64_60
8
res_8x64_80
60
0.18
52
134786
Use if NMULT must be 6
0.16
55
138292
Use if NMULT must be less than 8
60
0.03
62
134782
SUGGESTED DEFAULTBest 60% bandwidth mode, Adequate for most applications
80
0.7
43
139894
An 80% mode with NMULT=8
Configurations with NDELAY = 32,
These offer higher % bandwidth, but fewer resampling delays (-42dB inband jitter noise)
res_9x32_60
9
60
0.01
63
64700
Best 60% mode for interpolating by integer amounts
res_12x32_80
12
80
0.11
48
65610
Alternate if NMULT must be less than 15
res_15x32_80
15
0.05
58
66482
Best 80% mode, Very good for interpolating by integer amounts
Configurations with NDELAY = 16,
These are restricted to single or dual channel modes, very good for interpolating by integer amounts.
res_19x16_80
19
80
0.02
65
32988
Good filter, but only for single or dual channel modes
The %BW column reflects the percentage passband of the signal relative to the sample rate going into the resampler. The
resampler’s response is flat over this bandwidth to within the passband ripple indicated above. The image rejection is the amount
that the resampling images are rejected relative to the signal.
The filter taps for these configurations can be requested by e-mail from [email protected], or downloaded from
the web at www.graychip.com. The filter taps are available in two file formats. The first, <filter_name>.taps, is an ascii listing of
the taps. The second, <filter_name>.cmd, is a command file which could be used to write the filter taps into a GC4016 chip. The
commands are formatted as “write <address> <data_byte>”, where <address> and <data_byte> are unsigned hex integers.
 GRAYCHIP,INC.
- 60 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.8
DATA SHEET REV 1.0
GC4016 CONFIGURATION GENERATOR
A GC4016 configuration generator is available from Graychip in the form of a “C” program which can be obtained via email
or the web. The program is described below.
Usage: cmd4016 [-gc100 | -TI | -sim] <config_file>
where <config_file> is a file containing a list of control register settings of the form:
<parameter> <value>
print [gc100 | TI | sim ]
channel [0-3]
copy_channel [0-3]
cfir_coef <file>
pfir_coef <file>
res_coef <file>
where <parameter> is one of the gc4016 control register fields identified in the datasheet or quick reference guide.
Cmd4016 reads the file of control register settings and creates output files which summarize the control register values
needed to program the chip. The program creates config_file.tbl, which contains a summary of all of the control register settings
in a table format.
A second file is created which is in a format suitable for programming the chip. Three formats are supported: gc100, TI, and
sim. The format is either chosen on the command line, or is chosen using the “print [gc100 | TI | sim]” command. These options
create config_file.gc100, config_file.ti or config_file.sim output files.
The gc100 format is used by the gc100 test card interface. This format contains commands of the form:
write <address> <data>
or
write_block <starting_address> <data0> <data1> <data2> ...
The TI format is for use with the 4 bit addressing mode and contains commands of the form:
<address> <data>
The sim format is used by the “ifan” simulator and contains commands of the form:
writec <address> <data>
The “4_bit_address” mode is required in the TI mode, it is illegal in the gc100 or sim modes. A warning will be printed and
the appropriate 4_bit_address mode will be forced if this is violated. cmd4016 then checks the configuration for sanity and
calculates the gains throughout the chip.
Example configuration files can also be provided by email, or downloaded from the web.
 GRAYCHIP,INC.
- 61 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.9
DATA SHEET REV 1.0
EXAMPLE GSM APPLICATION
This section describes how to configure the chip to downconvert GSM (or EDGE) signals and meet the stringent GSM
processing specifications. The desired GSM Specifications are:
Table 29: Desired GSM Specifications
Specification
Value
Comment
Clock (CK)
69.333MHz =256*270.833KHz
Can be any value equal to 4NB, where N is the CIC decimation ratio and B is the GSM bit
rate of 270.833KHz
Input Sample Rate
Same as clock rate
Can be 1/2, 1/3, 1/4, ..., 1/N times the clock rate using the zero pad feature.
Input Format
A port, 14 Bit, 2’s Complement
Can be modified as desired
Spur free dynamic
range
> -110dB
The GC4016 NCO provides more than 115 dB of spur free dynamic range
Output Sample Rate
270.833KHz, complex data
Can be 2X, 4X, 8X or any multiple of this rate by changing the resampler ratios
Passband Width
160KHz
The filter response must be flat over this bandwidth to within the passband ripple
Passband Ripple
< 0.1dB peak to peak
To meet the GSM BER specification
Stopband Rejection
Set by GSM Specification
See the GSM adjacent channel and blocker mask in Figure 20
Output Format
Four serial streams, 24 bit I and Q data
Can be modified as desired
7.9.1 GSM Filter Response
Figure 20 shows the overall response using CFIR filter, cfir_68, and a set of PFIR coefficients tuned for GSM. The GSM out
of band rejection mask is also shown. The passband ripple is less than 0.1 dB (peak to peak). The passband ripple specification
for GSM is not directly specified. Instead it is specified in terms of bit error rate (BER). Trade offs between transition bandwidth
and passband ripple in order to optimize the BER can easily be made by reprogramming the PFIR filter.
Adjacent Band Rejection
Blocker Requirements
Figure 20. Frequency Response for the Example GSM Application
The PFIR coefficients are:
132 182 1 -309 -330 119 560 343 -424 -834 -278 696 960 149 -916 -1132 -212 1075 1514
421 -1518 -2374 -673 2529 3817 726 -4796 -6888 -650 12967 26898 32767
These coefficients are available by e-mail or over the web as pfir_gsm.taps. The PFIR_SUM for this set of coefficients is
96277.
 GRAYCHIP,INC.
- 62 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
7.9.2 Oversampling Using the Resampler
This example assumes the input sample rate is equal to 4*N*B, where N is the decimation in the CIC filter (See Section 5.6)
and B is the GSM bit rate (270.833 KHz). The outputs are one complex sample per bit (270.833 KHz) with the resampler set to
interpolate by unity (no interpolation). The output rate can be doubled or quadrupled as desired by changing the resampler ratio.
The resampler uses the configuration res_8x64_60. (See Section 7.7). This configuration introduces 0.03dB of passband ripple
and -62dB inband noise, neither being large enough to effect the BER.
The resampler ratio for one sample per bit (270.833KHz) is 0x04000000. It is 0x02000000 for two samples per bit
(541.666KHz) and it is 0x010000000 for four samples per bit (1.08333MHz).
7.9.3 Gain
The example configuration assumes a CIC decimation of N=64, which corresponds to an ADC clock rate (CK) of 69.333248
MHz. The values of SCALE and BIG_SCALE must be chosen to satisfy: ( SHIFT + SCALE + 6 × BIG_SCALE ) ≤ ( 62 – 5log2 N ) .
N is 64 and SHIFT is 4, so ( SCALE + 6 × BIG_SCALE ) ≤ 28 , which is satisfied by setting SCALE=4 and BIG_SCALE=4. If other
values of N are chosen, then SCALE and BIG_SCALE need to be modified as necessary. The overall gain is adjusted using
FINE_GAIN and FINAL_SHIFT. The overall gain is:

5 ( SHIFT + SCALE + 6 × BIG_SCALE – 62 )  COARSE CFIR_SUM PFIR_SUM FINE_GAIN
FINAL_SHIFT
1
RES_SUM
-)(2
- ) ( -------------------GAIN =  ( --------------------) ( ------------------)
)N 2
) ( -----------------) ( --------------------------------------------( 2
32768 × NDELAY
NZEROS + 1
65536
65536
1024


where NZEROS=0, N=64, SHIFT=4, SCALE=4, BIG_SCALE=4, COARSE=0, CFIR_SUM=83001, PFIR_SUM=96277,
RES_SUM=134782 and NDELAY=64. Because of the loss of 1/2 when converting real data to complex, the desired gain is 2.0.
This can be achieved by setting FINE_GAIN to 1070 and FINAL_SHIFT equal to 4.
7.9.4 GSM Configuration
The control register settings for this example are shown in table 30. It is assumed that output pin SO is tied to input pin SIA.
Table 30: Example GSM Configuration
Global Registers
Address
0
1
2
3
4
5
6
7
F8
00
00
-
27
DC
00
00
After configuration set address 0 to 08, then set address 5 to 5C
Paged Registers
Address
CH A
CH B
CH C
CH D
RES
OUT
16
17
18
19
20
Pages 0,1
Load CFIR coefficients: cfir_68.taps
Pages 2-5
Load PFIR coefficients: pfir_gsm.taps
Page 6
00
00
FREQ
Page 7
0C
77
00
Pages 8.9
Load CFIR coefficients: cfir_68.taps
Pages 10-13
Load PFIR coefficients pfir_gsm.taps
21
22
3F
70
3F
70
3F
70
3F
70
23
24
25
26
27
28
29
30
31
64
00
00
00
00
1D
-
2E
04
64
00
00
00
00
1D
-
2E
04
64
00
00
00
00
1D
-
2E
04
64
00
00
00
00
1D
-
2E
04
00
00
04
unused
20
22
Page 14
00
00
FREQ
Page 15
0C
77
00
unused
Pages 16,17
Load CFIR coefficients: cfir_68.taps
Pages 18-21
Load PFIR coefficients: pfir_gsm.taps
20
22
Page 22
00
00
FREQ
Page 23
0C
77
00
unused
Pages 24,25
Load CFIR coefficients: cfir_68.taps
Pages26-29
Load PFIR coefficients: pfir_gsm.taps
20
22
Page 30
00
00
FREQ
Page 31
0C
77
00
unused
Page 32-63
Load resampler coefficients: res_8x64_60.taps
20
22
Page 64
23
07
00
34
E4
00
00
00
unused
Page 65
00
00
00
04
00
00
00
04
00
00
00
04
00
Page 98
7F
40
28
01
E9
B0
E4
10
32
54
76
02
unused
 GRAYCHIP,INC.
- 63 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.10
DATA SHEET REV 1.0
EXAMPLE IS-136 DAMPS APPLICATION
This section describes how to configure the chip to downconvert DAMPS signals, apply the root-raised-cosine receive filter,
and output samples at two times, four times or eight times the DAMPS symbol rate. The desired DAMPS Specifications are:
Table 31: Desired DAMPS Specifications
Specification
Value
Comment
Clock (CK)
62.208MHz =8*320*24.3KHz
Can be any value equal to 8NB, where N is the CIC decimation ratio and B is the DAMPS
symbol rate of 24.3KHz
Input Sample Rate
Same as clock rate
Can be 1/2, 1/3, 1/4, ... 1/N of the clock rate by using the zero pad feature.
Input Format
A port, 14 Bit, 2’s Complement
Can be modified as desired
Spur free dynamic
range
> -110dB
The GC4016 NCO provides more than 115 dB of spur free dynamic range
Output Sample Rate
48.6KHz, complex data (2X)
Can be 4X, 8X or any multiple of the symbol rate by changing the resampler ratios
Passband Response
RRC with alpha=0.35
Matches the required receive filter
Stopband Rejection
Set by DAMPS Specification
See the DAMPS rejection mask in Figure 21
Output Format
Four serial streams, 24 bit I and Q data
Can be modified as desired
7.10.1 DAMPS Filter Response
Figure 21 shows the overall response using CFIR filter, cfir_68, and a set of RRC PFIR coefficients tuned for 2X oversampled
output data. The DAMPS out of band rejection mask is also shown.
Rejection Mask
Figure 21. Frequency Response for the Example DAMPS Application
The RRC PFIR coefficients are:
-12 -117 -113 8 127 112 -38 -175 -134 82 275 224 -71 -348 -307 61 398 286 -286 -761
-442 766 1954 1708 -660 -4042 -5641 -2533 6187 18177 28625 32767
These coefficients are available by e-mail or over the web as pfir_damps_2x.taps. The PFIR_SUM for this set of coefficients
is 119387.
7.10.2 Oversampling Using the Resampler
This example assumes the input sample rate is equal to 8*N*B, where N is the decimation in the CIC filter (See Section 5.6)
and B is the DAMPS symbol rate (24.3 KHz). The outputs are two complex samples per symbol (48.6 KHz) with the resampler
set to interpolate by unity (no interpolation). The output rate can be doubled or quadrupled as desired by changing the resampler
 GRAYCHIP,INC.
- 64 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
ratio. The resampler uses the configuration res_15x32_80. (See Section 7.7). This configuration introduces 0.05dB of passband
ripple and -58dB inband noise, neither being large enough to effect the BER.
The resampler ratio for two samples per symbol (48.6KHz) is 0x04000000, for four samples per symbol (97.2KHz) is
0x02000000, and for eight samples per symbol (194.4KHz) is 0x01000000.
7.10.3 Gain
The example configuration assumes a CIC decimation of N=320, which corresponds to an ADC clock rate (CK) of 62.208
MHz. The values of SCALE and BIG_SCALE must be chosen to satisfy: ( SHIFT + SCALE + 6 × BIG_SCALE ) ≤ ( 62 – 5log2 N ) .
N is 320 and SHIFT=4, so ( SCALE + 6 × BIG_SCALE ) ≤ 16.4 , which is satisfied by setting SCALE=4 and BIG_SCALE=2. If
other values of N are chosen, then SCALE and BIG_SCALE need to be modified as necessary. The overall gain is adjusted using
FINE_GAIN and FINAL_SHIFT. The overall gain is:

5 ( SHIFT + SCALE + 6 × BIG_SCALE – 62 )  COARSE CFIR_SUM PFIR_SUM FINE_GAIN
FINAL_SHIFT
1
RES_SUM GAIN =  ( --------------------)N 2
) ( ------------------) ( ------------------ ) ( -------------------) ( --------------------------------------------)(2
)
( 2
32768 × NDELAY
NZEROS + 1
65536
65536
1024


where NZEROS=0, N=320, SHIFT=4, SCALE=4, BIG_SCALE=2, COARSE=0, CFIR_SUM=83001, PFIR_SUM=119387,
RES_SUM=66554 and NDELAY=32. Because of the loss of 1/2 when converting real data to complex, the desired gain is 2.0.
This can be achieved by setting FINE_GAIN to 1147 and FINAL_SHIFT equal to 4.
7.10.4 DAMPS Configuration
The control register settings for this example are shown in table 32. It is assumed that output pin SO is tied to input pin SIA.
Table 32: Example DAMPS Configuration
Global Registers
Address
0
1
2
3
4
5
6
7
F8
00
00
-
27
DC
00
00
After configuration set address 0 to 08, then set address 5 to 5C
Paged Registers
Address
CH A
CH B
CH C
CH D
RES
OUT
16
17
18
19
20
21
Pages 0,1
Load CFIR coefficients: cfir_68.taps
Pages 2-5
Load PFIR coefficients: pfir_damps_2x.taps
Page 6
00
00
FREQ
Page 7
0C
77
00
Pages 8.9
Load CFIR coefficients: cfir_68.taps
Pages 10-13
Load PFIR coefficients pfir_damps_2x.taps
22
23
24
25
26
27
28
29
30
31
54
00
00
00
00
1D
-
7B
04
54
00
00
00
00
1D
-
7B
04
54
00
00
00
00
1D
-
7B
04
54
00
00
00
00
1D
-
7B
04
00
00
04
unused
20
22
3F
Page 14
00
00
FREQ
Page 15
0C
77
00
Pages 16,17
Load CFIR coefficients: cfir_68.taps
Pages 18-21
Load PFIR coefficients: pfir_damps_2x.taps
71
unused
20
22
3F
Page 22
00
00
FREQ
Page 23
0C
77
00
Pages 24,25
Load CFIR coefficients: cfir_68.taps
Pages26-29
Load PFIR coefficients: pfir_damps_2x.taps
71
unused
20
22
3F
Page 30
00
00
FREQ
Page 31
0C
77
00
Page 32-63
Load resampler coefficients: res_15_32_80.taps
71
unused
20
22
3F
71
Page 64
23
0E
00
34
E4
00
00
00
unused
Page 65
00
00
00
04
00
00
00
04
00
00
00
04
00
Page 98
7F
40
28
01
D9
B1
E4
10
32
54
76
02
unused
 GRAYCHIP,INC.
- 65 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.11
DATA SHEET REV 1.0
EXAMPLE IS-95 NB-CDMA APPLICATION
This section describes how to configure the chip to downconvert IS-95 NB-CDMA signals and output samples at two times
(2X), four times (4X) or eight times (8X) the IS-95 symbol rate of 1.2288MHz. The desired IS-95 Specifications are:
Table 33: Desired IS-95 NB-CDMA Specifications
Specification
Value
Comment
Clock (CK)
58.9824MHz =48*1.2288MHz
Can be any value greater than this, does not need to be a multiple of 1.2288MHz.
Input Sample Rate
Same as clock rate
Can be 1/2, 1/3, 1/4, ... 1/N times the clock rate using the zero pad feature
Input Format
A port, 14 Bit, 2’s Complement
Can be modified as desired
Spur free dynamic
range
> -110dB
The GC4016 NCO provides more than 115 dB of spur free dynamic range
Output Sample Rate
4.9152MHz, complex data (4X)
Can be 2X or 4X multiple of the symbol rate by changing the resampler ratio. Can be 8X
by using the dual channel mode.
Passband Width
615kHz
1.25MHz signal spacing
Passband Ripple
0.4 dB
Stopband Rejection
50 dB at 750kHz, 87dB at 900kHz offset
Far band rejection > 120dB
Output Format
Parallel output, 24 bit I and Q data
Can be modified as desired, allows simple AGC in an FPGA or ASIC to 4 or 6 bits.
7.11.1 IS-95 Filter Response
Figure 22 shows the overall response when the channel decimates to 1.5 times the baud rate. This uses the CFIR filter,
cfir_68, and the PFIR filter pfir_is95_1.5x. The passband ripple is less than 0.4 dB (peak to peak).
750 kHZ
900 kHz
615 kHz
-87 dB
Figure 22. Frequency Response for the Example IS95 Application
The PFIR coefficients are:
-4 -45 -166 -349 -452 -299 96 420 302 -225 -612 -321 471 881 267 -866 -1191-64 1461 1516 -388 -2351 -1826 1303 3804
2088 -3358 -6915 -2266 11185 26207 32767
These coefficients are available by e-mail or over the web as pfir_is95_1.5x.taps. The PFIR_SUM for this set of coefficients
is 89373.
 GRAYCHIP,INC.
- 66 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
7.11.2 Oversampling Using the Resampler
This example assumes the input sample rate is equal to 6*N*B, where N is the decimation in the CIC filter (See Section 5.6)
and B is the IS95 symbol rate (1.2288 MHz). The PFIR filter output rate is 1.5 samples per symbol (1.5X). The outputs are two
complex samples per symbol (2.4576 MHz) if the resampler is set to interpolate by 4/3. The output rate can be 4X or 8X as
desired by changing the resampler ratio. The resampler uses the configuration res_6x64_60. (See Section 7.7). This
configuration introduces -58dB inband noise, not enough to effect the BER. The spectral effect of the resampler is included in
Figure 22.
The resampler ratio for two samples per symbol (2.4576MHz) is 0x03000000, for four samples per symbol (4.9152MHz) it
is 0x01800000, and for eight samples per symbol (9.8304MHz) it is 0x00C00000. Note that the resampler can only handle two
channels at eight samples per symbol.
Sample rates other than 6*N*B can be accepted. To use these rates one needs to customize the PFIR filter to match the
bandwidth when the sample rate has been decimated by 4*N, and then the resampling ratio needs to be adjusted to match the
desired output rate. Call Graychip for details.
7.11.3 Gain
The example configuration assumes a CIC decimation of N=8, which corresponds to an ADC clock rate (CK) of 58.9824MHz.
The values of SCALE and BIG_SCALE must be chosen to satisfy: ( SHIFT + SCALE + 6 × BIG_SCALE ) ≤ ( 62 – 5log2 N ) . N is 8
and SHIFT=4, so ( SCALE + 6 × BIG_SCALE ) ≤ 43
, which is satisfied by setting SCALE=1 and BIG_SCALE=7. If other values
of N are chosen, then SCALE and BIG_SCALE need to be modified as necessary. The overall gain is adjusted using FINE_GAIN
and FINAL_SHIFT. The overall gain is:

5 ( SHIFT + SCALE + 6 × BIG_SCALE – 62 )  COARSE CFIR_SUM PFIR_SUM FINE_GAIN
FINAL_SHIFT
1
RES_SUM
-)(2
- ) ( -------------------GAIN =  ( --------------------) ( ------------------)
)N 2
) ( -----------------) ( --------------------------------------------( 2
32768 × NDELAY
NZEROS + 1
65536
65536
1024


where NZEROS=0, N=8, SHIFT=4, SCALE=1, BIG_SCALE=7, COARSE=0, CFIR_SUM=83001, PFIR_SUM=89373,
RES_SUM=134786 and NDELAY=64. Because of the loss of 1/2 when converting real data to complex, the desired gain is 2.0.
This can be achieved by setting FINE_GAIN to 1153 and FINAL_SHIFT equal to 4.
 GRAYCHIP,INC.
- 67 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
7.11.4 IS95 NB-CDMA Configuration
The control register settings for this example are shown in table 34. It is assumed that output pin SO is tied to input pin SIA.
Table 34: Example IS95 NB-CDMA Configuration
Global Registers
Address
0
1
2
3
4
5
6
7
F8
00
00
-
27
DC
00
00
After configuration set address 0 to 08, then set address 5 to 5C
Paged Registers
Address
CH A
CH B
CH C
CH D
RES
OUT
16
17
18
19
20
21
Pages 0,1
Load CFIR coefficients: cfir_68.taps
Pages 2-5
Load PFIR coefficients: pfir_is95_1.5x.taps
Page 6
00
00
FREQ
Page 7
0C
77
22
Pages 8.9
Load CFIR coefficients: cfir_68.taps
Pages 10-13
Load PFIR coefficients pfir_is95_1.5x.taps
22
23
24
25
26
27
28
29
30
31
79
00
00
00
00
1D
-
81
04
79
00
00
00
00
1D
-
81
04
79
00
00
00
00
1D
-
81
04
79
00
00
00
00
1D
-
81
04
00
80
01
unused
20
22
07
Page 14
00
00
FREQ
Page 15
0C
77
22
Pages 16,17
Load CFIR coefficients: cfir_68.taps
Pages 18-21
Load PFIR coefficients: pfir_is95_1.5x.taps
70
unused
20
22
07
Page 22
00
00
FREQ
Page 23
0C
77
22
Pages 24,25
Load CFIR coefficients: cfir_68.taps
Pages26-29
Load PFIR coefficients: pfir_is95_1.5x.taps
70
unused
20
22
07
70
Page 30
00
00
FREQ
Page 31
0C
77
22
unused
Page 32-63
Load resampler coefficients: res_6x64_60.taps
20
22
07
70
Page 64
23
05
00
14
E4
70
00
E4
unused
Page 65
00
00
80
01
00
00
80
01
00
00
80
01
00
Page 98
FF
40
6C
87
EF
00
E4
10
32
54
76
02
unused
7.12
UMTS WB-CDMA APPLICATION
This section describes how to configure the chip to downconvert UMTS WB-CDMA signals and output samples at two times
or four times the UMTS symbol rate. The desired UMTS Specifications are:
Table 35: Desired UMTS Specifications
Specification
Value
Comment
Clock (CK)
61.44MHz =16*3.84MHz
Can be any value greater than this, does not need to be a multiple of 3.84MHz.
Input Sample Rate
Same as clock rate
May be an integer division of the clock rate (30.72MHz, 20.48MHz, etc.)
Input Format
A port, 14 Bit, 2’s Complement
Can be modified as desired
Spur free dynamic
range
> -80dB
The GC4016 NCO provides more than 115 dB of spur free dynamic range
Output Sample Rate
15.36MSPS, complex data (4X)
Can be 2X or 4X multiple of the symbol rate by changing the resampler ratio. Can do two
channels at 2X if the clock rate is greater than 92.16MHz.
Passband Width
4.6848MHz
RRC filter with alpha=0.22 give a bandwidth of 1.22*3.84MHz
Passband Ripple
0.1 dB
Filter gives 0.05dB passband ripple
Stopband Rejection
60 dB at 5MHz, 80dB at > 5MHz offset
Rejection > 80dB at 2.6MHz offset
Output Format
Parallel output, 16 bit I and Q data
Can be modified as desired, allows simple AGC in an FPGA to 4 or 6 bits.
 GRAYCHIP,INC.
- 68 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
7.12.1 UMTS Filter Response
Figure 23 shows the overall response using specially tuned CFIR and PFIR filters., The PFIR filter is a specially optimized
root-raised-cosine 63 tap filter with an alpha of 0.22. The filter is designed to have less than -50dB of inter symbol interference
(ISI) noise.
2.5MHz
Figure 23. Frequency Response for the Example UMTS Application
These coefficients are available by e-mail or over the web as cfir_umts.taps and pfir_umts.taps. The CFIR_SUM for
cfir_umts.taps is 58827. The PFIR_SUM for pfir_umts.taps is 61821.
7.12.2 Oversampling Using the Resampler
This example assumes the input sample rate is equal to 4*N*B, where N is the decimation in the CIC filter (See Section 5.6)
and B is the UMTS symbol rate (3.84 MHz). The outputs are two complex samples per symbol (7.68MHz) with the resampler set
to interpolate by unity (no interpolation). The output rate can be doubled or quadrupled as desired by changing the resampler
ratio. The resampler uses the configuration res_8x64_60. (See Section 7.7). This configuration introduces 0.03dB of passband
ripple and -62dB inband noise, neither being large enough to effect UMTS demodulation.
The resampler ratio for two samples per symbol (7.68MHz) is 0x04000000 and for four samples per symbol (15.36MHz) it
is 0x02000000.
Sample rates other than 4*N*B can be accepted by adjusting the PFIR filter to match the desired RRC filter and by using the
resampler to adjust the final sample rate to be exactly 2X or 4X the symbol rate.
7.12.3 Gain
The example configuration assumes a CIC decimation of N=4, which corresponds to an ADC clock rate (CK) of 61.44 MHz.
The values of SCALE and BIG_SCALE must be chosen to satisfy: ( SHIFT + SCALE + 6 × BIG_SCALE ) ≤ ( 62 – 5log2 N ) . N is 4
and SHIFT=4, so ( SCALE + 6 × BIG_SCALE ) ≤ 48
, which is satisfied by setting SCALE=5 and BIG_SCALE=7, which are their
maximum values. If other values of N are chosen, then SCALE and BIG_SCALE need to be modified as necessary. The overall
gain is adjusted using COARSE, FINE_GAIN and FINAL_SHIFT. The overall gain is:

5 ( SHIFT + SCALE + 6 × BIG_SCALE – 62 )  COARSE CFIR_SUM PFIR_SUM FINE_GAIN
FINAL_SHIFT
1
RES_SUM
-)(2
- ) ( -------------------GAIN =  ( --------------------) ( ------------------)
)N 2
) ( -----------------) ( --------------------------------------------( 2
32768 × NDELAY
NZEROS + 1
65536
65536
1024


where NZEROS=0, N=4, SHIFT=4, SCALE=5, BIG_SCALE=7, COARSE=1, CFIR_SUM=58827, PFIR_SUM=61821,
 GRAYCHIP,INC.
- 69 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
RES_SUM=134782 and NDELAY=64. Because of the loss of 1/2 when converting real data to complex, the desired gain is 2.0.
This can be achieved by setting FINE_GAIN to 2352 and FINAL_SHIFT equal to 4.
7.12.4 UMTS Configuration
The control register settings for this example are shown in table 36. It is assumed that output pin SO is tied to input pin SIA.
Table 36: Example UMTS Configuration
Global Registers
Address
0
1
2
3
4
5
6
7
F8
00
00
-
27
DC
00
00
After configuration set address 0 to 08, then set address 5 to 5C
Paged Registers
Address
CH A
CH B
CH C
CH D
RES
OUT
16
17
18
19
20
Pages 0,1
Load CFIR coefficients: cfir_umts.taps
Pages 2-5
Load PFIR coefficients: pfir_umts.taps
Page 6
00
00
FREQ
Page 7
0C
77
22
Pages 8.9
Load CFIR coefficients: cfir_umts.taps
Pages 10-13
Load PFIR coefficients pfir_umts.taps
21
22
07
70
07
70
07
70
07
70
23
24
25
26
27
28
29
30
31
7D
30
10
06
00
1D
-
30
09
7D
50
10
06
00
1D
-
30
09
7D
30
10
00
00
1D
-
30
09
7D
50
10
00
00
1D
-
30
09
00
00
02
unused
20
22
Page 14
00
C0
FREQ
Page 15
0C
77
22
unused
Pages 16,17
Load CFIR coefficients: cfir_umts.taps
Pages 18-21
Load PFIR coefficients: pfir_umts.taps
20
22
Page 22
00
00
FREQ
Page 23
0C
77
22
unused
Pages 24,25
Load CFIR coefficients: cfir_umts.taps
Pages26-29
Load PFIR coefficients: pfir_umts.taps
20
22
Page 30
00
C0
FREQ
Page 31
0C
77
22
unused
Page 32-63
Load resampler coefficients: res_8x64_60.taps
20
22
Page 64
20
07
00
14
00
70
00
00
unused
Page 65
00
00
00
02
00
00
00
2
00
00
00
02
00
Page 98
FF
40
6C
80
29
01
00
10
32
54
76
02
unused
 GRAYCHIP,INC.
- 70 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
7.13
DATA SHEET REV 1.0
DIAGNOSTICS
The following four tables contain the diagnostic test configurations. To run the diagnostics, load the GC4016 with the
configuration, set address 0 to 00 in order to clear the resets, set address 5 to 00 to release the counter, wait for the checksum
to stabilize (approximately 220 clock cycles) and then read the checksum, which should match the expected value.
Table 37: Diagnostic Test 1, Expected Checksum = E0
GLOBAL REGISTERS
ADDRESS
VALUE
0
F0
1
00
2
00
3
00
4
2D
5
1C
6
FF
7
01
PAGED REGISTERS
PAGE
ADDRESS
16
17
20
21
22
23
24
25
26
27
28
29
30
31
CFIR COEFFICIENT PAGES: 0,1,8,9,16,17,24,25
0,8,16,24
FF
7F
00
00
00
18
19
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
1,9,17,25
00
00
00
00
00
00
PFIR COEFFICIENT PAGES: 2,3,4,5,10,11,12,13,18,19,20,21,26,27,28,29
2
3,4,5
FF
00
7F
00
00
00
FREQUENCY TUNING PAGES:
6
AA
AA
56
00
00
00
00
00
00
34
12
00
67
78
45
56
23
34
00
00
30
AA
AA
89
GENERAL CONTROL PAGES:
67
45
00
14
22
55
55
55
55
00
00
00
00
7,15,23,31
0C
77
55
50
55
07
70
78
00
01
RESAMPLER COEFFICIENT PAGES:
01
07
00
-
00
20
32
33-63
FF
00
07
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
64
53
46
00
14
E4
RESAMPLER CONTROL PAGE:
50
00
E4
65
00
00
00
04
00
RESAMPLER RATIO PAGE:
00
00
04
00
00
00
04
00
00
00
04
D9
OUTPUT PAGE:
E4
21
65
87
00
02
98
00
A1
6C
C1
00
00
00
00
00
00
00
00
00
43
.
Table 38: Diagnostic Test 2, Expected Checksum = 4C
PAGED REGISTERS
PAGE
ADDRESS
16
17
18
19
20
CFIR COEFFICIENT PAGES: 0,1,8,9,16,17,24,25
0,8,16,24
1,9,17,25
42
2A
00
27
2C
FD
01
39
93
91
21
22
23
24
25
26
27
28
29
30
31
01
41
6C
FE
0B
F8
A0
F5
2E
FE
CA
10
PFIR COEFFICIENT PAGES: 2,3,4,5,10,11,12,13,18,19,20,21,26,27,28,29
2
1C
00
AD
00
0D
00
4C
FF
2A
FF
51
00
9D
01
2B
01
58
B2
FD
04
CA
EE
FE
04
25
9E
02
FF
8B
C3
03
F9
D5
9A
00
F9
A1
69
FC
00
CD
8A
FB
08
5
A1
09
23
FREQUENCY TUNING PAGES:
00
6C
F2
57
EE
23
FF
88
22
E2
46
4A
56
70
71
00
00
00
07
00
-
00
08
3
4
EA
D5
FE
FF
6
14
AA
55
AA
55
67
78
45
56
23
34
00
00
22
30
55
AA
55
AA
89
9A
67
78
45
56
00
00
GENERAL CONTROL PAGES:
7,15,23,31
0C
77
55
50
55
0F
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
00
04
00
00
00
04
43
65
87
00
02
32
FF
07
00
00
RESAMPLER COEFFICIENT PAGES:
00
00
00
00
00
00
33-63
00
00
00
00
00
00
00
00
00
RESAMPLER CONTROL PAGE:
64
53
46
00
14
E4
50
00
E4
RESAMPLER RATIO PAGE:
65
00
00
00
04
00
00
98
00
A1
6C
C1
D9
00
 GRAYCHIP,INC.
00
04
OUTPUT PAGE:
E4
21
- 71 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
.
Table 39: Diagnostic Test 3, Expected Checksum = EF
PAGED REGISTERS
PAGE
ADDRESS
16
17
20
21
22
23
24
25
26
27
28
29
30
31
CFIR COEFFICIENT PAGES: 0,1,8,9,16,17,24,25
0,8,16,24
0C
00
A3
FF
C2
18
19
FF
24
03
03
05
07
FB
B3
EB
48
F2
1,9,17,25
8C
20
F7
60
FF
7F
PFIR COEFFICIENT PAGES: 2,3,4,5,10,11,12,13,18,19,20,21,26,27,28,29
2
3
02
8B
00
FF
01
F8
00
FF
F5
BF
FF
00
E9
9B
FF
00
FE
43
FF
FF
2D
89
00
FE
2B
1A
00
00
D0
43
FF
02
4
5
66
50
01
04
A7
6A
FD
0E
6A
4D
FC
03
F8
16
00
EB
BD
7E
05
EA
6A
AE
02
19
70
D5
F9
5E
35
FF
F8
7F
FREQUENCY TUNING PAGES:
6
AA
AA
89
67
45
FF
94
45
78
89
56
67
FF
FF
30
AA
AA
56
GENERAL CONTROL PAGES:
94
78
FF
14
22
55
55
55
55
7,15,23,31
0C
77
55
50
55
1F
70
6A
00
00
RESAMPLER COEFFICIENT PAGES:
00
07
00
-
55
05
32
33
F7
F4
0F
0F
F6
F3
0F
0F
F6
F2
0F
0F
F6
F1
0F
0F
F5
F1
0F
0F
F5
F0
0F
0F
F4
EF
0F
0F
F4
EE
0F
0F
34
35
EE
E8
0F
0F
ED
E8
0F
0F
EC
E7
0F
0F
EB
E7
0F
0F
EB
E6
0F
0F
EA
E6
0F
0F
E9
E6
0F
0F
E9
E6
0F
0F
36
37
E5
E7
0F
0F
E5
E8
0F
0F
E5
E9
0F
0F
E6
EA
0F
0F
E6
EB
0F
0F
E6
EC
0F
0F
E6
ED
0F
0F
E7
EE
0F
0F
38
39
F0
FF
0F
0F
F1
02
0F
00
F3
04
0F
00
F5
07
0F
00
F6
9
0F
00
F8
0C
0F
00
FB
0F
0F
00
FD
12
0F
00
40
41
15
30
00
00
19
34
00
00
1C
37
00
00
1F
3B
00
00
22
3E
00
00
26
42
00
00
29
45
00
00
2D
48
00
00
42
43
4C
62
00
00
4F
65
00
00
52
67
00
00
55
69
00
00
58
6A
00
00
5B
6C
00
00
5E
6D
00
00
60
6E
00
00
44
45
6E
69
00
00
6F
67
00
00
6F
65
00
00
6F
62
00
00
6E
5F
00
00
6D
5B
00
00
6C
57
00
00
6B
53
00
00
46
47
4F
1D
00
00
4A
15
00
00
44
0D
00
00
3F
05
00
00
39
FC
00
0F
32
F3
00
0F
2C
EA
00
0F
24
E0
00
0F
48
49
D7
83
0F
0F
CD
78
0F
0F
C3
6D
0F
0F
B8
62
0F
0F
AE
57
0F
0F
A3
4C
0F
0F
98
41
0F
0F
8D
36
0F
0F
50
51
2C
E2
0F
0E
22
DA
0F
0E
17
D3
0F
0E
0E
CD
0F
0E
4
C7
0F
0E
FB
C1
0E
0E
F2
BD
0E
0E
EA
B9
0E
0E
52
53
B5
B7
0E
0E
B3
BB
0E
0E
B1
C1
0E
0E
B0
C7
0E
0E
AF
CE
0E
0E
B0
D6
0E
0E
B1
E0
0E
0E
B4
EA
0E
0E
54
55
F5
79
0E
0F
02
8E
0F
0F
10
A5
0F
0F
1E
BD
0F
0F
2E
D6
0F
0F
3F
EF
0F
0F
51
0A
0F
00
65
26
0F
00
56
57
43
4D
00
01
61
72
00
01
80
98
00
01
A0
BE
00
01
C1
E5
00
01
E3
0D
00
02
05
35
01
02
29
5D
01
02
58
59
86
D8
02
03
B0
02
02
04
D9
2D
02
04
03
57
03
04
2E
81
03
04
58
AB
03
04
83
D4
03
04
AD
FD
03
04
60
61
25
50
05
06
4D
72
05
06
74
92
05
06
9B
B1
05
06
C1
CF
05
06
E6
EC
05
06
0A
08
06
07
2E
23
06
07
62
63
3C
D0
07
07
53
DC
07
07
6A
E6
07
07
7F
EE
07
07
92
F5
07
07
A4
FA
07
07
B4
FD
07
07
C3
FF
07
07
64
53
7
00
14
E4
RESAMPLER CONTROL PAGE:
50
00
E4
65
56
34
12
04
89
RESAMPLER RATIO PAGE:
67
45
02
00
89
67
05
90
68
47
04
65
87
00
02
98
 GRAYCHIP,INC.
00
A1
6C
C1
19
OUTPUT PAGE:
00
E4
21
- 72 -
43
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
.
Table 40: Diagnostic Test 4, Expected Checksum = 18
PAGED REGISTERS
PAGE
ADDRESS
16
17
20
21
22
23
24
25
26
27
28
29
30
31
CFIR COEFFICIENT PAGES: 0,1,8,9,16,17,24,25
0,8,16,24
2F
00
9A
FF
60
18
19
FB
06
FB
CF
05
AB
0C
FA
F8
B0
E1
1,9,17,25
0F
00
CB
51
FF
7F
PFIR COEFFICIENT PAGES: 2,3,4,5,10,11,12,13,18,19,20,21,26,27,28,29
2
3
E0
0D
FF
00
16
3A
FF
FF
62
49
00
01
E8
BD
FF
FE
B6
95
FF
00
A5
91
00
00
39
4B
FF
FE
8A
41
00
02
4
5
38
0E
FE
F9
49
AA
00
06
BD
6D
01
FD
9D
CD
FC
FA
BB
45
03
0F
C0
B6
FD
E6
2B
A8
FF
20
72
FF
04
7F
70
70
64
64
30
50
00
00
02
02
07
07
00
00
-
00
00
04
04
FREQUENCY TUNING PAGES:
6
00
00
67
45
23
01
45
56
23
34
01
02
30
00
C0
71
56
GENERAL CONTROL PAGES: 7,15,23,31
34
02
7F
7F
14
22
00
00
C0
00
67
71
7,23
15,31
0C
0C
77
77
55
55
50
50
55
55
32
FF
07
00
00
RESAMPLER COEFFICIENT PAGES:
00
00
00
00
00
00
00
00
00
00
00
00
33-63
00
00
00
00
00
00
00
00
00
RESAMPLER CONTROL PAGE:
00
00
00
00
00
00
00
64
53
46
00
14
E4
50
00
E4
RESAMPLER RATIO PAGE:
65
00
00
00
04
00
00
00
00
00
04
00
00
00
04
98
0F
A8
28
C1
19
00
43
65
87
00
02
00
04
OUTPUT PAGE:
E4
21
Electronic versions of these tables are available by email or from the web as diag_tests.txt.
7.14
OUTPUT TEST CONFIGURATION
The output test takes advantage of the tag modes. To perform the output tests, first configure the chip according to the
desired mode, and then change the configuration so that:
TAG_EN=1 and FLUSH_SYNC=7 (always) in all four channels.
TAG_EN turns on the tag mode, and FLUSH_SYNC clears the channels, but allows the channels to continue to operate and
output zeroes. Then follow the suggested initialization sequence described in Section 3.12. This will force the output words to be
zeroes with the tag bits (see addresses 23-26 in the output page) in the LSBs.
 GRAYCHIP,INC.
- 73 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 MULTI-STANDARD QUAD DDC CHIP
DATA SHEET REV 1.0
Graychip reserves the right to make changes in circuit design and/or specifications at any time without notice. The
user is cautioned to verify that datasheets are current before placing orders. Information provided by Graychip is
believed to be accurate and reliable. No responsibility is assumed by Graychip for its use, nor for any infringement
of patents or other rights of third parties which may arise from its use. No license is granted by implication or
otherwise under any patent rights of Graychip.
 GRAYCHIP,INC.
- 74 -
August 27, 2001
This document contains information which may be changed at any time without notice
GC4016 REGISTER ASSIGNMENT QUICK REFERENCE GUIDE.
Address
Global
0
Global Reset
1
Status
2
Page
3
Checksum
4
General Syncs
5
Count Sync
6
Counter Byte 0
CNT[0:7]
00
7
Counter Byte 1
CNT[7:15]
00
0,1
16-31
Name
CFIR Coef
7(MSB)
6
GLOBAL_
RESET
OUT_BLK_
RESET
5
4
PAD_RESET
ZERO
3
RESAMPLER_
RESET
CHECK_DONE
EDGE_WRITE
2
1
CK_2X_EN
RES_QOV
CK_2X_TEST
RES_IOV
C
H
LVDS
4_BIT_ADDR
ONE_SHOT
OS_MODE
DIAG_SYNC
COUNTER_SYNC
C
H
C
H
-
DITHER_SYNC
ZPAD_EN
ZPAD_SYNC
-
FLUSH_SYNC
A
Control
20 (0x14)
Dec and Flush
21 (0x15)
Dec Byte 0
22 (0x16)
Dec Byte 1
-
23 (0x17)
CIC Scale
-
24 (0x18)
SplitIQ
25 (0x19)
CFIR
26 (0x1A)
PFIR
27 (0x1B)
Input
28 (0x1C)
Peak Control
29 (0x1D)
Peak Count
PEAK_COUNT
30 (0x1E)
Fine Gain Byte 0
FINE_GAIN[0:7]
31 (0x1F)
Fine Gain Byte 1
77
NCO_SYNC
22
20
22
07
MIX20B
QONLY
TEST
DEC[8:11]
BIG_SCALE
IONLY
SPLITIQ
MSB_POL
-
-
-
PEAK_SELECT
IN_FMT
PEAK_MODE
79
NEG_CTL
COARSE
-
70
SCALE
QDLY_CFIR
NO_SYM_CFIR
00
IDLY_PFIR
QDLY_PFIR
NO_SYM_PFIR
00
PEAK_THRESHOLD
-
00
IDLY_CFIR
AB_SEL
INPUT_SEL
00
PEAK_SYNC
1D
read only
00
FINE_GAIN[8:13]
10-13
16-31
PFIR Coef
32 PFIR Coefficients for Channel B, Load LSBs in even addresses, MSBs in odd addresses
14
16,17
PHASE
16 bit channel B phase, LSBs in 16, MSBs in 17, PHASE=216P/2π
04
0000
32 bit channel B tuning frequency, LSBs in 18, MSBs in 20, FREQ = 232F/FCK
15
16-31
Channel Control See above
16,17
16-31
CFIR Coef
11 CFIR Coefficients for Channel C, Load LSBs in even addresses, MSBs in odd addresses
18-21
16-31
PFIR Coef
32 PFIR Coefficients for Channel C, Load LSBs in even addresses, MSBs in odd addresses
22
16,17
PHASE
16 bit channel C phase, LSBs in 16, MSBs in 17, PHASE=216P/2 π
00000000
0000
32 bit channel C tuning frequency, LSBs in 18, MSBs in 20, FREQ = 2 32F/FCK
23
16-31
Channel Control See above
24,25
16-31
CFIR Coef
11 CFIR Coefficients for Channel D, Load LSBs in even addresses, MSBs in odd addresses
26-29
16-31
PFIR Coef
32 PFIR Coefficients for Channel D, Load LSBs in even addresses, MSBs in odd addresses
30
16,17
PHASE
16 bit channel D phase, LSBs in 16, MSBs in 17, PHASE=216P/2 π
16-31
0C
DEC_SYNC
GAIN_SYNC
11 CFIR Coefficients for Channel B, Load LSBs in even addresses, MSBs in odd addresses
31
SHIFT
PHASE_SYNC
DEC[0:7]
CFIR Coef
18,19,20,21 FREQ
0000
-
16-31
D
D0 then 50
00000000
NZEROS
8,9
18,19,20,21 FREQ
00
27
DIAG_SOURCE
-
Zero Pad
C
00
11 CFIR Coefficients for Channel A, Load LSBs in even addresses, MSBs in odd addresses
19 (0x13)
18,19,20,21 FREQ
F8 then 08
See below
read only
OUTPUT_SYNC
COUNT_TEST
NCO Sync
B
READY
CHECKSUM[0:7]
18 (0x12)
Page
CK_LOSS_
DETECT
A3
C
32 PFIR Coefficients for Channel A, Load LSBs in even addresses, MSBs in odd addresses
2,3,4,5
16-31
PFIR Coef
H
16 bit channel A phase, LSBs in 16, MSBs in 17, PHASE=216P/2π
A
6
16,17
PHASE
N
32 bit channel A tuning frequency, LSBs in 18, MSBs in 21, FREQ = 232F/FCK
18,19,20,21 FREQ
N
CH_RESET
USE_SHIFT
E
7
16 (0x10) Channel Reset
L
FREQ_SYNC
17 (0x11) Frequency Sync
A Channel
MISSED
PAGE[0:6]
0(LSB)
Suggested
Default
Page
00000000
0000
32 bit channel D tuning frequency, LSBs in 18, MSBs in 20, FREQ = 2 32F/FCK
00000000
Channel Control See above
Power up initialization: (Assumes SO is tied to SIA) (1) Set address 0 to F8; (2) Write to all registers and coefficients; (3) Set address 5 to D0;
(4) Set address 0 to 08; (5) Set address 5 to 50.
SYNC MODE
SYNC SOURCE
0,1
off (never asserted)
2
SIA
3
SIB
4
ONE_SHOT
5
TC (terminal count of internal counter)
6,7
on (always active)
REV 1.0
August 27, 2001
GC4016 REGISTER ASSIGNMENT QUICK REFERENCE GUIDE
Page
R 32-63
E
64
S
(0x40)
A
M
P
L
E
R
Address
16-31
Filter Taps
7(MSB)
N-Channels
-
17 (0x11)
N-Multiplies
-
18 (0x12)
Filter Select
19 (0x13)
Final Shift
20 (0x14)
Channel Map
21 (0x15)
Add To
22 (0x16)
Clock Divide
23 (0x17)
Ratio Map
16-19
96,97
6
5
4
3
2
1
0(LSB)
Suggested
Default
Resampler Coefficients, 8 LSBs in even addresses, 4 MSBs in odd addresses. (Must be loaded in the blocks 16-23 and 24-31)
16 (0x10)
65
(0x41)
O
98
U (0x62)
T
P
U
T
Name
RES_SYNC
NF=(NFILTER-1)
NO_SYM_RES
FILTER_SEL_2
TAG_22
CHAN_MAP_D
FILTER_SEL_0
00
CHAN_MAP_A
E4
FINAL_SHIFT
CHAN_MAP_C
RATIO_SYNC
46
FILTER_SEL_1
ROUND (12B,16B, 20B,24B)
-
23
NM=(NMULT-1)
FILTER_SEL_3
-
Bypass
NC =(NCHAN-1)
CHAN_MAP_B
(must be 0)
ADD_C_TO_D
35
ADD_B_TO_C
ADD_A_TO_B
70
RES_CLK_DIV (The resampler clock rate is 2*F CK/(RES_CLK_DIV+1))
RATIO_MAP_3
RATIO_MAP_2
00
RATIO_MAP_1
RATIO_MAP_0
00
Res Ratio 0
RATIO_0, Resampler ratio 0. RATIO = 226 (Resampler input sample rate)/(Resampler output sample rate)
04000000
20-23
Res Ratio 1
RATIO_1, Resampler ratio 1.
04000000
24-27
Res Ratio 2
RATIO_2, Resampler ratio 2.
04000000
28-31
Res Ratio 3
RATIO_3, Resampler ratio 3.
04000000
16-31
Output Data
See Tables 4,5, and 6 for details
EN_PAR
EN_P3
EN_P2
EN_P1
EN_P0
EN_SFS
EN_RDY
EN_SCK
RDY_WIDTH
TAG_EN
INV_SFS
INV_RDY
INV_SCK
REAL_ONLY
MASTER
PARALLEL
NIBBLE
LINK
16 (0x10)
Tristate Controls
17 (0x11)
Output Format
18 (0x12)
Output Mode
19 (0x13)
Frame Control
20 (0x14)
Word Sizes
21 (0x15)
Clock Control
22 (0x16)
Serial Mux
23 (0x17)
Output Tag A
TAG_AQ
TAG_AI
10
24 (0x18)
Output Tag B
TAG_BQ
TAG_BI
32
25 (0x19)
Output Tag C
TAG_CQ
TAG_CI
54
26 (0x1A)
Output Tag D
TAG_DQ
TAG_DI
27 (0x1B)
Revision
28 (0x1C)
Miscellaneous
OUT_BLK_SYNC
REVERSE_IQ
OUTPUT_MODE
SFS_MODE
See Table 3
FRAME_LENGTH
BLOCK_SIZE
BITS_PER_WORD
OUTPUT_ORDER
WORDS_PER_FRAME
NSERIAL
SMUX_3
SCK_RATE
SMUX_2
SMUX_1
SMUX_0
76
Mask Revision Number
read only
-
EN_SO
EN_4_FS
02
.
Global Syncs (Addresses 4 and 5)
Channel Syncs (Pages 7, 15, 23 and 31)
Sync
Value
Description
Sync
Value
Description
DIAG_SYNC
7 (always)
Only used during diagnostics
PHASE_SYNC
7 (always)
Use phase settings as they are loaded
OUTPUT_SYNC
4 (OS)
The SO output is used during initialization
FREQ_SYNC
7 (always)
Use frequency settings as they are loaded
COUNTER_SYNC
4 (OS)
Sync counter with one shot pulse
NCO_SYNC
2 (SIA)
Sync the phase accumulator during
initialization. Set to SIB for frequncy hopping.
DITHER_SYNC
0 (never)
or 2 (SIA)
Can free run except during diagnostics, or
reset during initialization
ZPAD_SYNC
2 (SIA)
Sync the zero pad circuit during initialization
DEC_SYNC
2 (SIA)
Sync the channel decimation during
initialization
Outout Circuit Sync (Page 98)
OUT_BLK_SYNC
2 (SIA)
Sync the output block during initialization
Resampler syncs (Page 64)
RES_SYNC
2 (SIA)
Sync the Resampler during initialization
FLUSH_SYNC
2 (SIA)
Flush the channels during initialization
RATIO_SYNC
7(always)
Set to always except when synchronously
changing ratios
GAIN_SYNC
7 (always)
Use fine gain settings as they are loaded
PEAK_SYNC
5 (TC)
Periodically capture peak count data

5 ( SHIFT + SCALE + 6 × BIG_SCALE – 62 )  COARSE CFIR_SUM PFIR_SUM FINE_GAIN
FINAL_SHIFT
1
RES_SUM GAIN =  ( --------------------)N 2
) ( ------------------) ( ------------------ ) ( -------------------) ( --------------------------------------------)(2
)
( 2
32768 × NDELAY
NZEROS + 1
65536
65536
1024


REV 1.0
August 27, 2001
PACKAGE OPTION ADDENDUM
www.ti.com
4-Mar-2005
PACKAGING INFORMATION
Orderable Device
Status (1)
Package
Type
Package
Drawing
GC4016-PB
ACTIVE
BGA
GJZ
Pins Package Eco Plan (2)
Qty
160
126
None
Lead/Ball Finish
SNAGCU
MSL Peak Temp (3)
Level-3-220C-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 - May not be currently available - please check http://www.ti.com/productcontent for the latest availability information and additional
product content details.
None: Not yet available Lead (Pb-Free).
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.
Green (RoHS & no Sb/Br): TI defines "Green" to mean "Pb-Free" and in addition, uses package materials that do not contain halogens,
including bromine (Br) or antimony (Sb) above 0.1% of total product weight.
(3)
MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDECindustry 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
Telephony
www.ti.com/telephony
Video & Imaging
www.ti.com/video
Wireless
www.ti.com/wireless
Mailing Address:
Texas Instruments
Post Office Box 655303 Dallas, Texas 75265
Copyright  2005, Texas Instruments Incorporated