1/31/02AAdded instructions for programming ADM1024, Sect 3.6. Added channel number coding
for Channel Tag in output format, Section 4.2, per KB Case 972. Added detailed installation
instructions for Option 102, Sect 2.3. Removed Virtex−E FPGA Appendix—data is in FPGA
Design Kit, Part #49530−250. Added Input Offset Error Specifications, per KB case 1016.
6/14/02A.1Added notes on Real mode output from GC4016s in DDR modes, Sect 4.2, per KBcase 1073.
9/26/02A.2Added note on Reset hold time, Sect 3.7, per KBcase 961.
per KBcase 1138. Corrected FPGA mating connector ERNI part #, Sect 2.4.5, per KBcase 1129.
2/24/03A.4Corrected Sync/Gate & FPGA connector part numbers, Sect 2.4.3 and 2.4.5, per KBcase 1155.
3/4/03A.5Corrected DDR Bypass mode descriptions, Sect 4.2.1 and 4.2.2.
5/21/03BAdded Option 105 information.
6/6/03B.1Corrected Configuration Data Register to Write Only, Sect 3.4.
7/15/03B.2Added VME baseboards, Sect 1.1, added Support Software Sect 1.11, corrected Model 4205
Added comment that input channel data goes only to associated BIFO for Bypass modes,
Sect 4.2.1 & 4.2.2, per KBcase 1153. Moved Timing & Sync to new Chapter 5.
6/25/04B.4Added note about more samples than trigger length , Sect 3.11, per KBcase 1161.
1/14/05CUpdated Power Specifications, Sect 1.12.
Pentek warrants that all products manufactured by Pentek conform to published Pentek specifications and are free from defects in mate−
rials and workmanship for a period of one year from the date of delivery when used under normal operating conditions and within the
service conditions for which they were furnished.
The obligation of Pentek arising from a warranty claim shall be limited to repairing or at its option, replacing without charge, any product
that in Pentek’s sole opinion proves to be defective within the scope of the warranty. Pentek must be notified in writing of the defect or
nonconformity within the warranty period and the affected product returned to Pentek within thirty days after discovery of such defect or
nonconformity.
Buyer shall prepay shipping charges, taxes, duties and insurance for products returned to Pentek for warranty service. Pentek shall pay
for the return of products to buyer except for products returned from another country.
Pentek shall have no responsibility for any defect or damage caused by improper installation, unauthorized modification, misuse, neglect,
inadequate maintenance, or accident, or for any product that has been repaired or altered by anyone other than Pentek or its authorized
representatives.
The warranty described above is buyer’s sole and exclusive remedy and no other warranty, whether written or oral, is expressed or
implied. Pentek specifically disclaims fitness for a particular purpose. Under no circumstances shall Pentek be liable for any direct, indi−
rect, special, incidental, or consequential damages, expenses, losses or delays (including loss of profits) based on contract, tort, or any
other legal theory.
Appendix B, AD6644 Datasheet, is the copyrighted property of Analog Devices, Inc., Norwood MA
Appendix C, Graychip GC4016 Datasheet, is the copyrighted property of Graychip, Inc., Palo Alto, CA
Appendix D, ADM1024 Datasheet, is the copyrighted property of Analog Devices, Inc., Norwood MA.
Trademarks
Pentek, GateFlow, ReadyFlow, and VIM are registered trademarks or trademarks of Pentek, Inc.
VxWorks is a registered trademark of Wind River Systems, Inc. Xilinx and Virtex are registered trademarks or trademarks of Xilinx, Inc.
The Models 6230 and 6231 are general purpose, narrowband digital receiver VIM®
(Velocity Interface Mezzanine) modules that perform frequency down conversion,
low−pass filtering, and decimation of the digitized input signal. The Model 6230 is a
32−channel VIM−4 module that features four 14−bit, 65−MHz A/D converters, and can
be configured with up to 32 channels of narrowband receivers. The Model 6231 is a
16−channel VIM−2 module, with two 14−bit, 65−MHz A/D converters, and up to 16
channels of narrowband receivers.
The Models 6230 and 6231 attach directly to VIM−compatible baseboards, including the
Pentek Models 4205, and 4290 through 4295 Quad DSP (Digital Signal Processing)
boards.
1.2Features
Four 65−MHz, 14−bit A/D converters (Model 6230)
Two 65−MHz, 14−bit A/D converters (Model 6231)
32 channels of narrowband digital receivers (Model 6230)
16 channels of narrowband digital receivers (Model 6231)
All receivers can select any of the A/D inputs
DC to 90−MHz input range (standard models)
300−kHz to 150−MHz input range (Option 105)
DC to 32−MHz center frequency tuning with 0.02−Hz resolution
3.6−kHz to 1.6−MHz output bandwidths
Front panel clock and sync bus can synchronize multiple boards
Direct connection to each VIM processor with no shared bus bottlenecks
Compliant with VIM module specification
In this manual, all specifications and user instructions are presented in reference to
Note
the Model 6230. Where a difference exists, the corresponding information for the
Model 6231 is included in brackets “[...]”.
Rev.: B.4
Page 10
Page 10Pentek Model 6230/6231 Operating Manual
1.3Analog/Digital Conversion
The Model 6230 [Model 6231] accepts four [two] analog RF inputs on front panel SMA
connectors in the range of DC to 90 MHz to support direct IF undersampling. Each
input signal is buffered by an Analog Devices OPA642 amplifier. Low−pass anti−
aliasing filters for each input signal may be individually enabled or bypassed by on−
board jumpers. With Option 105, the amplifier, filters, and bypass jumpers are
replaced by an RF transformer, and the RF input range is 300 kHz to 150 MHz.
An Analog Devices AD6644 14−bit, 65−MHz A/D converter then digitizes each of the
analog inputs. The A/D converter clock can be driven from an internal 64−MHz crys−
tal oscillator, from an external sample clock supplied through a front panel SMA con−
nector, or from the front panel sync bus.
1.4Digital Receivers
The Model 6230 [Model 6231] includes eight [four] Graychip GC4016 quad narrowband
digital receiver (DDR) chips. Each GC4016 DDR accepts four [two] 14−bit parallel
inputs from the AD6644 A/D converters. A crossbar switch inside each GC4016 allows
each of the 32 [16] receiver channels on the board to independently select any of the
A/D inputs for flexible switching.
The maximum input sampling rate, f
includes four independently tunable receiver channels capable of tuning throughout
the DC to f
/2 range, with output bandwidths ranging from 0.8 · fs/N (for the standard
s
80% FIR filter), where the decimation factor, N, ranges from 32 to 16,384. For an input
sampling clock of 64 MHz, this output bandwidth range becomes approximately
3.1 kHz to 1.6 MHz. Each processor on the VIM baseboard can control all programma−
ble registers on its two associated GC4016s.
1.5Digital Interfaces
The Model 6230 [Model 6231] contains two [one] Xilinx® Virtex®−E XCV300 FPGAs
(field programmable gate arrays). (These are optionally replaceable by other Virtex
devices in the series up to XCV600.) The FPGAs are factory programmed to implement
the standard data formatting, clocking, and control functions specified in this docu−
ment. Each GC4016 DDR delivers real or complex serial data streams into one FPGA,
where data is formatted and multiplexed for delivery across the VIM interface into the
32−bit parallel Bi−FIFOs (BIFOs) on the VIM baseboard. The AD6644 digital outputs
are also connected directly to each FPGA so that wideband A/D data can be delivered
directly to the DSP board, bypassing the digital receivers.
The user can re−program the Virtex−E FPGAs from the VIM baseboard processors.
This allows the user to implement his own algorithms for special timing requirements
and for pre−processing of AD6644 or GC4016 output data before sending it to the pro−
cessor BIFOs. Refer to Section 1.10, FPGA Configuration, for additional information
about the gate array configurations.
, for the GC4016 is 80 MHz. Each receiver
s
Rev.: B.4
Page 11
Pentek Model 6230/6231 Operating ManualPage 11
1.6Timing and Synchronization
A VIM baseboard processor can access all board control and status registers to control
synchronization, gating, triggering, and clocking functions. All VIM baseboard pro−
cessors can generate sync, gate, and trigger signals for distribution on the front panel
LVDS (low−voltage differential signal) sync bus. This sync bus includes sample clock,
gate, and sync signals. It allows one Model 6230/6231 to act as a bus Master, driving
these signals out to a front panel flat cable using LVDS differential signaling. Addi−
tional sync lines on the bus allow synchronization of the local oscillator phase, fre−
quency switching, decimating filter phase, and BIFO data collection on multiple
modules. Up to seven slave 6230/6231 modules can be driven from the bus Master,
supporting synchronous sampling and sync functions across all connected boards. If
the Model 6230/6231 is a bus Master, any of the VIM baseboard processors can create a
sync individually (one at a time) by toggling a bit in a register.
In addition to the LVDS sync bus, the Model 6230/6231 can receive two external input
signals: one TTL sync and one TTL gate or trigger. Refer to Chapter 5, Gate/Sync
Description, for additional information about the use and programming of gate, clock,
and sync signals.
1.7Interrupts
The Model 6230/6231 has several maskable interrupt sources. Interrupts may be gen−
erated to any of the VIM baseboard processors by the A/D converter overload outputs,
transitions on the gate or sync signals, clock loss, or a programmable over−temperature
or a faulty power supply voltage.
The board’s Voltage/Temperature Monitor, an ADM1024, provides constant monitor−
ing of critical voltages and temperatures on the Model 6230/6231 PCB. This device is
programmable for voltage and temperature limits. If the voltage/temperature fall out−
side of the set limits, an interrupt can be generated.
1.8VIM Interface
The FPGA outputs are connected directly through the VIM mezzanine interface to the
32−bit synchronous BIFOs on the VIM baseboard, where they are buffered for efficient
block transfers to the processors. Each processor on the VIM baseboard can control all
programmable registers on the associated GC4016s, as well as control and initiate sync
bus functions.
Rev.: B.4
Page 12
Page 12Pentek Model 6230/6231 Operating Manual
1.9Block Diagrams
The following are block diagrams of the Models 6230 and 6231 digital receivers.
Sample
Clock In
CLOCK
GENERATOR/
DIVIDER
Clock
& Sync
Bus
TTL
Sync &
Gate
SYNCHRONIZATION
INTERRUPTS
AND CONTROL
Front Panel I/O
Model 6230
VIM Processor Board
XTL
OSC.
A/D Clock
GC4016
NARROW−
BAND
DIG. RCVR
16
Control
4−CH
RF In
AMPLIFIER
LOW PASS
FILTER
AD6644
14−BIT A/D
14141414
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
AMPLIFIER
LOW PASS
FILTER
AD6644
14−BIT A/D
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
RF In
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
I&QI&QI&QI&QI&QI&QI&QI&Q
Virtex−E FPGAVirtex−E FPGA
32
I & QI & Q
PROC
BI−
FIFO
A
3232
PROC
B
I & QI & Q
BI−
FIFO
32
ControlControlControl
Figure 1−1: Model 6230 Block Diagram
AMPLIFIER
LOW PASS
14−BIT A/D
PROC
C
RF In
FILTER
AD6644
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
323232
BI−
FIFO
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
PROC
AMPLIFIER
LOW PASS
FILTER
AD6644
14−BIT A/D
NARROW−
BAND
DIG. RCVR
D
32
RF In
GC4016
BI−
FIFO
4−CH
Front
16
Panel
I/O
Rev.: B.4
Sample
Clock In
CLOCK
GENERATOR/
DIVIDER
Clock
& Sync
Bus
TTL
Sync &
Gate
SYNCHRONIZATION
INTERRUPTS
AND CONTROL
Model 6231
VIM Processor Board
Figure 1−2: Model 6231 Block Diagram
XTL
OSC.
A/D Clock
GC4016
NARROW−
BAND
4−CH
DIG. RCVR
Control
PROC
A (or C)
RF In
AMPLIFIER
LOW PA SS
FILTER
AD6644
14−BIT A/D
1414
GC4016
NARROW−
BAND
DIG. RCVR
4−CH
GC4016
NARROW−
BAND
DIG. RCVR
I&QI&QI&QI&Q
Virtex−E FPGA
32
Control
I & Q
BI−
FIFO
32
4−CH
PROC
B (or D)
AMPLIFIER
LOW PA SS
FILTER
AD6644
14−BIT A/D
GC4016
NARROW−
BAND
DIG. RCVR
BI−
FIFO
32
RF In
4−CH
32
I & Q
16
Front Panel I/O
(optional)
Page 13
Pentek Model 6230/6231 Operating ManualPage 13
1.9Block Diagrams (continued)
The following are block diagrams of the Models 6230 and 6231 with Option 105.
Sample
Clock In
CLOCK
GENERATOR/
DIVIDER
Clock
& Sync
Bus
TTL
Sync &
Gate
SYNCHRONIZATION
INTERRUPTS
AND CONTROL
Front Panel I/O
Model 6230
VIM Processor Board
XTL
OSC.
A/D Clock
BAND
DIG. RC VR
GC4016
NARROW−
4−CH
RF In
RF
TRANSFORMER
AD6644
14−BIT A/D
14
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
RF InRF InRF In
RF
TRANSFORMER
AD6644
14−BIT A/D
141414
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
GC4016
NARROW−
BAND
DIG. RC VR
TRANSFORMER
AD6644
14−BIT A/D
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
RF
4−CH
GC4016
NARROW−
BAND
DIG. RC VR
I&QI&QI&QI&QI&QI&QI&QI&Q
16
Control
PROC
A
3232
Virtex−E FPGAVirtex−E FPGA
32
I & QI & Q
FIFO
BI−
PROC
B
I & QI & Q
BI−
FIFO
32
ControlControlControl
PROC
C
323232
BI−
FIFO
Figure 1−3: Model 6230 Block Diagram, with Option 105
RF
TRANSFORMER
AD6644
14−BIT A/D
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
PROC
FIFO
D
32
BI−
4−CH
Front
16
Panel
I/O
Clock
& Sync
Bus
TTL
Sync &
Gate
Sample
Clock In
CLOCK
GENERATOR/
DIVIDER
SYNCHRONIZATION
INTERRUPTS
AND CONTROL
XTL
OSC.
A/D Clock
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
RF In
RF
TRANSFORMER
AD6644
14−BIT A/D
14
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
RF In
RF
TRANSFORMER
AD6644
14−BIT A/D
14
GC4016
NARROW−
BAND
4−CH
DIG. RC VR
I&QI&QI&QI&Q
16
Model 6231
VIM Processor Board
Control
PROC
A (or C)
Virtex−E FPGA
32
Control
I & Q
BI−
FIFO
32
PROC
B (or D)
32
I & Q
BI−
FIFO
32
Front Panel I/O
(optional)
Figure 1−4: Model 6231 Block Diagram, with Option 105
Rev.: B.4
Page 14
Page 14Pentek Model 6230/6231 Operating Manual
1.9.1Channels
The following defines the different uses of the term ‘Channel’ in this manual
(refer also to the block diagrams on the prior two pages).
Input Channels — There are four input channels on the Model 6230
[two input channels on the Model 6231], one for each analog RF input
and A/D converter. These are identified as CH1, CH2, CH3, and CH4
for the 6230 [CH1 and CH2 for the 6231]. Each input channel may be
independently selected by any of the receiver channels (below).
Receiver Channels — Each VIM baseboard processor has access to eight
receiver channels, four channels in each of the two Graychip GC4016
DDRs associated with that processor. These are identified as DSP CH1
through DSP CH8, for each processor. Any one of the A/D input
channels (above) may be connected independently to each receiver
channel. Each receiver channel may be individually selected for output
to the VIM Bi−FIFO by the associated processor.
Processor Channels — There are four processor channels on the Model
6230 [two processor channels on the 6231], one for each processor on the
VIM baseboard. These are identified as Processors A, B, C, and D for the
6230 [Processors A and B, or C and D, for the 6231, depending on the
mezzanine position that the module is installed into on the VIM
baseboard]. Each processor channel is associated with (and provides
control for) two GC4106s on the module, and one Bi−FIFO on the VIM
baseboard. Two processor channels are associated with each FPGA
(Processor A controls FPGA1, Processor C controls FPGA2).
Rev.: B.4
Page 15
Pentek Model 6230/6231 Operating ManualPage 15
1.10FPGA Configuration
The standard FPGAs on the Model 6230/6231 are the Xilinx Virtex−E XCV300E; the
Virtex−E XCV600E models are available as Option 600. The baseline functionality con−
sumes about 60% of the XCV300E or 30% of the XCV600E.
All data from the AD6644 A/D converters and the Graychip GC4016 DDRs pass
through these FPGAs before being fed to the VIM baseboard BIFOs. The Model 6230/
6231 is shipped with a default set of logic functions for the FPGAs, on JTAG−program−
mable serial EEPROMs. Each FPGA has it’s own EEPROM to facilitate two different
programmings. At power−up, this set of default functions is loaded into the FPGAs.
The FPGAs can be configured in several different ways:
•The default method is configuration loading from separate 4−Mbit configuration
EEPROMs. This is the power−up mode of the board. Configuration reload may
also be forced by Processors A and C (see Section 3.3).
•The second method, to facilitate development and debugging, is by serial download
to the FPGAs using a Xilinx download cable. This method will replace the factory−
programmed configuration that loads at power up from the EEPROMs. In this
mode, the FPGAs are chained to accept one download stream, so two different or
identical programs can be strung together in the download bitstream. This method
is volatile and will exist only until the power is turned off.
•The third method is byte−wide upload from Processors A and C. Processor A
configures the first FPGA, and Processor C the second, by writing the configuration
data to the FPGA Configuration Data Register (see Section 3.4). For a Virtex
XCV600E FPGA, approximately 512K bytes on the baseboard are required to hold
the configuration for each FPGA. This method is volatile and will exist only until
the power is turned off.
•The last method is to use the serial EEPROM configuration method, then overwrite
the default configuration by reprogramming each EEPROM from the JTAG
interface. This overwrites the default configuration by reprogramming the
EEPROMs. After turning power off, the FPGAs will power up with this new
configuration instead of the default configuration.
NOTE:
This method will permanently overwrite the default configuration supplied
by Pentek. The default configuration is supplied with the available FPGA
Design Kit so that it can be restored if necessary.
Pentek has available GateFlow™ FPGA Design Kits that provide resources for the user
to modify the programmable logic functions for the Virtex−E FPGA. This allows the
user to implement his own algorithms for special timing requirements and for pre−
processing of AD6644 or GC4016 output data. Pentek offers this capability as a sepa−
rate development package, Model 4953 − Option 230 for the 6230, or Model 4953 −
Option 231 for the Model 6231. Contact Pentek at (201) 818−5900 for details about this
package.
Rev.: B.4
Page 16
Page 16Pentek Model 6230/6231 Operating Manual
1.10FPGA Configuration (continued)
1.10.1Model 6230
Some spare pins on the Model 6230 FPGAs are cross−connected to facilitate
user programming or future functionality. The spare pins are broken up
into the following groups:
•The first group of 38 spares is available on either FPGA model. Twelve
pins are connected to a serial port on Processors A and C through a
quick switch, but are disabled in the default configuration (see Note
below). This optional connection allows a path for the two serial ports to
the front panel connector if desired. These spare pins split into two
groups of 19, and are cross−connected so that identical programming
and I/O pin numbers in both FPGAs is possible.
•The second group of 44 spares exists only in the XCV600E model. They
are divided into two groups of 22 pins and cross−connected.
•The third group of 16 pins and two quickswitch enables from each FPGA
go to the front panel FPGA connector (see Section 2.4.5). These are
connected with either FPGA model (XCV300E or XCV600E), and can be
used as either inputs or outputs. The quickswitches and 25−ohm series
resistors provide some over−voltage and short protection to the FPGA
I/O pins. See Section 3.22 for description of this I/O register.
When the Model 6230 is attached to a Pentek VIM baseboard, the FPGAs
provide serial port connectivity as shown in the following illustration.
Virtex−E FPGA 1Virtex−E FPGA 2
Control
Quick
Switch
Model 6230
VIM Baseboard
Serial Ports
Control
Quick
Switch
A0A1B0B1C0C1D0D1
The control signals needed to enable connection of serial ports A1 and C1 to
the FPGAs are disabled in the default Model 6230 FPGA configuration. If
Note
needed, these connections will have to be optionally programmed into the
FPGA logic.
Rev.: B.4
Figure 1−5: Model 6230 FPGA Interconnectivity
Page 17
Pentek Model 6230/6231 Operating ManualPage 17
1.10FPGA Configuration (continued)
1.10.2Model 6231
All spare pins on the Model 6231 FPGA are brought to the front panel FPGA
connector (see Section 2.4.5). These pins are connected with either FPGA
model (XCV300E or XCV600E), and can be used as either inputs or outputs.
See Section 3.22 for description of the associated I/O register.
When the Model 6231 is attached to a Pentek VIM baseboard, the FPGA
provides serial port connectivity as shown in the following illustration.
Virtex−E FPGA
Model 6231
VIM Baseboard
Serial Ports
A0
(or C0)A1(or C1)B0(or D0)B1(or D1)
Figure 1−6: Model 6231 FPGA Interconnectivity
1.11Board Support Software
Pentek’s Model 4999 ReadyFlow® Board Support Libraries allow high−level program−
ming to speed development tasks. Refer to the ReadyFlow software documentation for
the Model 6230/6231 (Pentek part #801.62300) for further description of these capabili−
ties.
Pentek’s Model 4996 VxWorks
workstation platforms. In addition to the feature set provided by the driver, source
code is included, allowing users to modify software functionality. Refer to the
VxWorks Driver documentation for the Model 6230/6231 (Pentek part #803.62300) for
further description of these capabilities.
®
Driver allows high−level programming for various
Rev.: B.4
Page 18
Page 18Pentek Model 6230/6231 Operating Manual
1.12Specifications
Front Panel Connectors
Analog Inputs:Four [two] female SMA connectors (one per A/D converter)
Sample Clock Input:One female SMA connector
Sync/Gate Bus:One 26−pin connector, with four gates, one sync, and one
clock input/output LVDS signals, plus one sync and
one gate input TTL signals
TTL Sync/Gate:One 4−pin header, with one sync and one gate TTL inputs
FPGA Input/Output:
Model 6230:
Model 6231, Option 102:
Analog Signal Inputs
Quantity:Four [two], via front panel SMA connectors
Input Type:Single−ended, non−inverting
Coupling:DC
Input Impedance:50
Full Scale Input:Standard: 2 V
Input Offset Error:± 0.6% full scale range @ 25° C
One 50−pin connector, with 32 FPGA input/output pins,
20 mA maximum load per pin
One 50−pin connector, with 24 FPGA input/output pins,
20 mA maximum load per pin
Ω
(± 1.0 V)
p−p
Option 105: 1 V
(± 0.5 V)
p−p
± 1.6% full scale range over operating temperature range
Analog Input Conditioning
Analog Input Amplifiers
Quantity:Four [two] (enclosed in a shielded cover)
Device:Analog Devices OPA642
Gain:One (unity), standard
3 dB Bandwidth:90 MHz (based on gain of one)
Analog Input Filters
Quantity:Four [two] (enclosed in a shielded cover)
Type:Fixed frequency low−pass, 7−pole, LC
Passband:DC to 26 MHz (±3.0 dB ripple)
Stopband Attenuation:>70 dB @ 60 MHz
Bypass:May be bypassed by on−board jumper selection
Analog Input Conditioning
Analog Input Transformers
Quantity:Four [two] (enclosed in a shielded cover)
Type:Mini−Circuits ADT4−5WT
3 dB Passband:300 kHz to 500 MHz
Input Return Loss:8.72 dB min., 31.13 dB max.
Bypass:None
(without Option 105)
(with Option 105)
Rev.: B.4
Page 19
Pentek Model 6230/6231 Operating ManualPage 19
1.12Specifications (continued)
Analog/Digital Converters
Quantity:Four [two] (enclosed in a shielded cover)
Device:Analog Devices AD6644 (see Appendix B)
Sampling Rate:15 MHz to 65 MHz
Resolution:14 bits
Coupling:DC
Clock Source:Onboard crystal oscillator, external clock, or LVDS clock
(software selectable)
External Clock Input
Voltage Range:1 V
Type:Square or Sine Wave
Duty Cycle:45% to 55%
Frequency:15 MHz to 65 MHz
Impedance:50
Sample Rate Control
Internal Clock:64 MHz
External Clock:15 MHz to 65 MHz
Sample Rate Divider:Divisible by 1, 2, or 4 (internal or external source)
(minimum), 5 V
P−P
Ω, AC coupled
(maximum)
P−P
Gates
Quantity:Four [two], one per VIM baseboard BIFO
Polarity:Programmable
Routing:Each VIM baseboard processor can create its own gate
Gate Disable:Each gate can be disabled from its processor
(BIFO writes default to enable)
Digital Receivers
Quantity:Eight [four] quad receiver chips (32 [16] receiver channels)
Device:Graychip GC4016 (see Appendix C)
Decimation:32 to 16,384
Data Source:All A/D outputs are connected to each GC4016
Clock Source:A/D clock
Sync:Maskable inside each GC4016 chip
(All receiver channels directed to the same VIM base−
board BIFO must be synced at the GC4016 output)
Output:Serial data
(Serial output rate must be same as input clock rate)
Bypass Mode:Data from the A/D converters can be written directly into the
VIM baseboard BIFOs (bypassing the GC4016s), at a sam−
ple rate equal to the A/D clock decimated by 1 or any
even value between 2 and 4096
Rev.: B.4
Page 20
Page 20Pentek Model 6230/6231 Operating Manual
1.12Specifications (continued)
Field−Programmable Gate Arrays
Quantity:Two [one], one for Processors A & B, one for Processors C & D
Device:Xilinx Virtex−E XCV300E (standard);
Option 600 – Xilinx Virtex−E XCV600E
Programming:Factory programmed by Pentek
(contact Pentek for user programming information)
Estimated Power Dissipation:
+2.5V Digital Supply(DC/DC):
Graychip GC4016s:
+5V Analog Supply
(LC Filtered +5V):
AD6644 A/D Converters:
OPA642 Amplifiers:25 mA * 4 = 0.10 A25 mA * 2 = 0.05 A
AD8138 Amplifiers:24 mA * 4 = 0.10 A24 mA * 2 = 0.05 A
−5V Analog Supply
(Linear from −12V):
OPA642 Amplifiers:
AD8138 Amplifiers:24 mA * 4 = 96 mA24 mA * 2 = 48 mA
+1.8V Digital Supply
(DC/DC from 5V):
XCV300E FPGAs w/base circuitry:1.3 W0.65 W
XCV300E FPGAs w/heavy use:5.6 W2.8 W
XCV600E FPGAs w/heavy use:11.8 W5.91 W
+3.3V Digital Supply
(DC/DC from 12V):
XCV300E FPGAs w/base circuitry:1.2 W0.6 W
XCV300E FPGAs w/heavy use:1.3 W0.65 W
XCV600E FPGAs w/heavy use:1.9 W0.95 W
Total Board Power:
XCV300E FPGAs w/base circuitry:15 W7.55 W
XCV300E FPGAs w/heavy use:19.4 W9.75 W
XCV600E FPGAs w/heavy use:25.2 W13.15 W
(Notes: Estimated at 80MHz — power consumption is less at 65 MHz;
FPGA power consumption will vary based on usage)
Model 6230Model 6231
(at 80MHz and minimum decimation)
8 * 187mA = 1.5A4 * 187mA = 0.75A
1.5A * 2.5V = 3.75W
3.75W/.93 =4.0 W1.875W/.93 =2.0 W
260 mA * 4 = 1.04 A260 mA * 2 = 0.52 A
1.24 A * 5V =6.2 W0.62 A * 5V = 3.1 W
25 mA * 4 = 100 mA25 mA * 2 = 50 mA
196 mA * 12V =2.3 W98 mA * 12V = 1.2 W
0.75A * 2.5V = 1.875W
______________
Physical
Dimensions:6230 VIM−4 Module6231 VIM−2 Module
Height:228.6 mm (9.00 in)114.3 mm (4.50 in)
Depth: 82.5 mm (3.25 in)82.5 mm (3.25 in)
Width: 20.3 mm (0.80 in) 20.3 mm (0.80 in)
Weight:260.8 grams (9.2 oz) 138.9 grams (4.9 oz)
Environmental
Operating Temperature:0° to 50°C
Storage Temperature:−20° to 90°C
Relative Humidity:0 to 95% non−condensing
Rev.: C
Page 21
Pentek Model 6230/6231 Operating ManualPage 21
Chapter 2: Installation and Connections
2.1Inspection
After unpacking, inspect the unit carefully for possible damage to connectors or com−
ponents. If any damage is discovered, contact Pentek immediately at (201) 818−5900.
Please save the shipping container and packing material in case reshipment is required.
2.2Jumper Block Settings
The Model 6230 [Model 6231] PCB has ten [six] jumper blocks that the user can set.
With Option 105, the PCB has only two jumper blocks. These jumpers are described in
the following subsections. Assembly drawings of the Model 6230 and 6231 PCBs are
provided in Figures 2−1 and 2−2, showing the locations of these jumpers. Note that
most components are normally hidden under a heat sink/shield. To remove this
shield, remove ten [five] Phillips screws that hold the heat shield, from the solder side
of the board. Be sure to replace this shield after making any jumper changes.
On boards without Option 105, jumper blocks JB1 to JB8 [JB1 to JB4]
allow you to bypass the analog input filters (see Figures 1−1 and 1−2)—these
jumpers are not present with Option 105. For each input channel, you must
set two jumper blocks, one on either side of the input filter. Each jumper
block has three pins (as illustration in Table 2−1, below). The arrangement of
the pins on each jumper block is the same, however, the jumper blocks are
oriented differently on the PCB. Note the location of pin 1 of each jumper
block (Figures 2−1 and 2−2). For each jumper block, set the jumper on pins 1
− 2 to bypass the analog filter, or set the jumper on pins 2 − 3 to route the
analog input through the filter (default setting).
Pins 1 − 2 = Bypass the analog filter
Pins 2 − 3 = Use the analog filter *
Jumper Position
* (factory default setting)
For a given input channel, make sure to set both jumper blocks to the same
Note
pin positions.
Rev.: B.4
Page 22
Page 22Pentek Model 6230/6231 Operating Manual
2.2Jumper Block Settings (continued)
2.2.2FPGA Configuration Data Source Jumper
Jumper JB9 selects the source of the Virtex FPGA configuration data down−
load. The FPGA can select its configuration data either from an on−board
Serial EEPROM or from a serial download (X−Checker cable), depending on
the setting of this jumper. The following table shows the jumper settings for
this FPGA configuration data source jumper. This jumper block has three
pins, similar to the Filter Bypass jumpers (see illustration in Table 2−1, on
prior page). (Refer to Virtex Config Register, Section 3.3, for further infor−
mation on reconfiguring the Virtex FPGAs.)
Table 2−2: FPGA Configuration Data Source Jumper
Jumper JB9 PositionFPGA Data Source
Pins 1 − 2Serial Download
Removed, or pins 2 − 3 *On−board Serial EEPROM
* Factory Default Setting
2.2.3External TTL Inputs Select Jumpers
Jumper block JB10 selects the input connector to use as the source of the
external TTL SYNC or TTL GATE inputs. Each TTL signal can be input from
either the 26−pin sync/gate connector (see Section 2.4.3) or the 4−pin sync/
gate header (see Section 2.4.4) on the front panel. The following table shows
the jumper settings for this jumper block.
Table 2−3: External TTL Inputs Select Jumpers
Jumper JB10 PositionExternal TTL Gate/Sync Inputs
Pins 1 − 2
Pins 3 − 4
Installed
Removed
Installed
Removed
1
2
1
2
1 − Factory Default Setting on Model 6231
2 − Factory Default Setting on Model 6230
TTL SYNC from 26−pin Sync/Gate Connector
TTL SYNC from 4−pin Sync/Gate/Trig Header
TTL GATE from 26−pin Sync/Gate Connector
TTL GATE from 4−pin Sync/Gate/Trig Header
Rev.: B.4
Page 23
Pentek Model 6230/6231 Operating ManualPage 23
2.2Jumper Block Settings (continued)
Upper
J1
JB2
Front
Panel
Rear
Mounting
Holes
(a)
J2
J3
JB1
JB4
JB3
JB9
JB9
JB10
Filter
Bypass
Jumpers
(
Shield
removed
JB6
JB5
)
(b)
Figure 2−1: Model 6230 PCB Assembly, Component Side
JB8
J4
JB7
Rev.: B.4
Page 24
Page 24Pentek Model 6230/6231 Operating Manual
t
2.2Jumper Block Settings (continued)
Front
J1
JB2
Panel
Rear
Mounting
Hole
JB1
(a)
JB4
JB10
JB9
J2
JB3
Figure 2−2: Model 6231 PCB Assembly, Component Side
Fron
Panel
Panel Mounting Holes
JB9
JB10
Filter
Bypass
Jumpers
Shield
(
removed
)
J2
Figure 2−3: Model 6231 Option 102 PCB Assembly, Component Side
Rev.: B.4
J13
Standoff Mounting Holes
J12
J1
Page 25
Pentek Model 6230/6231 Operating ManualPage 25
2.3Installing the Model 6230/6231 on a VIM Baseboard
This section provides instructions for installing the Model 6230 VIM−4 module or the
Model 6231 VIM−2 module on a VIM−compatible baseboard. Pentek’s VIM base−
boards ship with two blank panels and four VIM interface connectors where you can
install optional VIM modules, such as the Model 6230 or Model 6231. An illustration of
a typical Pentek VIM baseboard is provided in Figure 2−4 on the following page.
Perform all assembly steps at an antistatic workstation.
CAUTION
Tools required for all procedures:
•#1 Phillips screwdriver
•Flat−blade screwdriver (blade width 5/16−inch or less)
The Models 6230 and 6231 are shipped in different configurations. The installation
instructions vary according to the configuration you receive from Pentek, as follows:
If you have ordered a Model 6230, or a Model 6231 without Option 102, for
installation on your VIM baseboard, you have received a single VIM module that
you must mount on your existing VIM baseboard.
Refer to Section 2.3.1 for the instructions for this configuration.
If you have ordered a Model 6231 with Option 102 for installation on your VIM
baseboard, you have received an assembly of a Model 6231 PCB and an Option 102
PCB that you must disassemble and mount on your existing VIM baseboard.
Refer to Section 2.3.2 for the instructions for this configuration.
Note
All required mounting hardware and front panels are included with your shipment.
Be sure to follow all instructions in the order presented.
Rev.: B.4
Page 26
Page 26Pentek Model 6230/6231 Operating Manual
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
(a)
A
(b)
(c)
Top Position
Mounting Holes
Bottom Position
Mounting Holes
B
C
(f)
(e)
Top Position
VIM Connectors
Bottom Position
VIM Connectors
(d)
Figure 2−4: Typical VIM Baseboard − Connectors & Mounting Holes
Rev.: B.4
D
Page 27
Pentek Model 6230/6231 Operating ManualPage 27
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.1Installing Model 6230, or Model 6231 without Option 102
This section provides instructions for installing a Model 6230 VIM−4 or
Model 6231 VIM−2 (without Option 102) on your existing VIM baseboard.
The Model 6230 or 6231 is shipped as an assembled unit and must be disas−
sembled for installation.
1) Remove both front panels from the Model 6230 VIM−4 module [single
front panel from the Model 6231 VIM−2 module] by removing the two
countersunk Phillips screws from each panel (see following illustration).
(Model 6230 illustrated, with lower panel removed)
2) Remove the (red) shipping brackets that hold the front panel(s) to the VIM
module by removing the pan−head Phillips shipping bracket screws from
the solder side of the VIM PCB (see illustration above). Set these screws
aside, as they are used to secure the VIM module to the VIM baseboard.
(The shipping brackets may be discarded, or saved to store the panels on
the VIM module if it is removed from the VIM baseboard.)
3) At the rear of the Model 6230 PCB, on the component side, are two nylon
spacers [one spacer on the Model 6231 VIM−2 PCB] (see Figure 2−6 on the
next page). Remove the nylon screws that are threaded into the top of the
spacers from the component side. Set these screws aside, as they are used
to secure the VIM module to the VIM baseboard.
Rev.: B.4
Page 28
Page 28Pentek Model 6230/6231 Operating Manual
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.1Installing Model 6230, or Model 6231 w/o Option 102 (continued)
Note
Nylon Spacer
Nylon Screw
VIM module
Figure 2−6: VIM Module Nylon Spacer
The nylon spacers must be in the holes at the REAR of the VIM connectors,
farthest from the front panel (as illustrated above, and indicated at locations (a)
and (b) in Figures 2−1 and 2−2). If either spacer is in the front hole, reposition it
using the nylon screw on the solder side of the module.
Rev.: B.4
4) For the Model 6230, remove both blank panel inserts from the VIM
baseboard, by removing the two countersunk Phillips screws from each
insert (see illustration below). For the Model 6231, remove one blank
panel insert, at the VIM position you wish to install the VIM−2 module.
Blank Panel Screws
3
2
1
0
3
3
3
2
1
0
210
210
Figure 2−7: VIM Baseboard Blank Panel Screws
Page 29
Pentek Model 6230/6231 Operating ManualPage 29
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.1Installing Model 6230, or Model 6231 w/o Option 102 (continued)
5) With the VIM baseboard’s component side (the side with the VIM
connectors) facing up, align the four VIM connectors on the Model 6230
VIM−4 module (J1, J2, J3, and J4) with the four VIM connectors on the
baseboard. For the Model 6231, align the two mezzanine connectors on
the VIM−2 module (J1 and J2) with two of the VIM connectors on the
baseboard—the Model 6231 VIM−2 module may be installed in either the
top or bottom mezzanine position on the VIM baseboard. See Figures 2−1
and 2−2 for location of the Models 6230 and 6231 VIM connectors, and see
Figure 2−4 for location of the VIM connectors on a typical VIM baseboard.
6) GENTLY but firmly, press down on the VIM module opposite the
connectors, to fully seat the module’s connectors into the baseboard’s. If
you meet with significant resistance, check the connector alignment.
NOTE:
Misalignment can cause bent pins or break connector
housings, so NEVER APPLY EXCESSIVE FORCE.
7) After seating the connectors, secure the front of the VIM module to the
VIM baseboard by screwing four [two for the Model 6231] pan−head
Phillips screws through the holes at the front of the VIM module into the
threaded holes in baseboard’s panel brackets (indicated at positions (a),
(b), (c), and (d) on Figure 2−4).
8) Turn the assembly over, such that the VIM module is on the work surface
and the solder side of the VIM baseboard is facing up. Secure the
baseboard to the nylon spacer(s) on the VIM module, using the two nylon
screws removed earlier [one for the Model 6231], through the rear
mounting hole(s) on the baseboard (indicated at positions (e) and (f) on
Figure 2−4).
9) Attach the VIM module’s front panel(s) to the baseboard, by screwing the
countersunk Phillips screws through the holes of each VIM panel into the
threaded holes in the front of the baseboard’s panel brackets. Note that
the Model 6230 panels fit only in their proper locations on the baseboard.
The Model 6230/6231 installation is complete.
Rev.: B.4
Page 30
Page 30Pentek Model 6230/6231 Operating Manual
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.2Installing Model 6231 with Option 102
The Model 6231 with Option 102 is shipped as an assembled unit and must
be disassembled for installation on your baseboard. A separate Option 102
front panel assembly is also provided.
1) Remove both front panels from the Model 6231 VIM−2 module and the
Option 102 module by removing the two countersunk Phillips screws
from each panel (see following illustration).
Standoff Screws (5)
Standoffs
VIM Panel Screws
Shipping Bracket Screws (2)
Stacking Connectors (2)
VIM Panel
Shipping Brackets
(Side View)
Figure 2−8: Model 6231 with Option 102 Shipping Assembly
(Shown with Option 102 panel removed)
2) Remove the (red) shipping brackets that hold the front panels to both
boards by removing the two pan−head Phillips shipping bracket screws
from the solder side of the Option 102 board (see illustration above).
(The shipping brackets may be discarded, or saved to store the panels if
they are removed.)
3) Remove and set aside the five pan−head Phillips standoff screws from the
top of the Option 102 board (see figure above).
4) CAREFULLY, remove the Option 102 board from the stacking connectors
(see figure above), taking care to not bend the pins.
5) GENTLY, remove and set aside both stacking connectors from the Model
6231 PCB. Do not remove the five (black) metal standoffs from this PCB.
Be careful when removing the stacking connectors, as they are fragile
CAUTION
Rev.: B.4
and can easily be damaged.
Page 31
Pentek Model 6230/6231 Operating ManualPage 31
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.2Installing Model 6231 with Option 102 (continued)
6) A nylon spacer is installed on the component side of the Model 6231 PCB,
between the VIM connectors. Remove and set aside the nylon screw that
is threaded into the top of the spacer (see following illustration).
Note
Nylon Spacer
Nylon Screw
VIM module
Figure 2−9: VIM Module Nylon Spacer
The nylon spacer must be installed in the hole at the REAR of the VIM
connectors, farthest from the front panel (as shown in the figure above).
If the spacer is installed in the front hole, reposition it using the nylon screw
on the solder side of the module.
7) Remove one blank panel insert from your VIM baseboard, at the VIM
position you wish to install the Model 6231 module, by removing the two
countersunk Phillips screws from that blank panel insert (see following
illustration).
Blank Panel Screws
321
3
2
1
0
0
210
1
0
3
Figure 2−10: VIM Baseboard Blank Panel Screws
3
2
Rev.: B.4
Page 32
Page 32Pentek Model 6230/6231 Operating Manual
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.2Installing Model 6231 with Option 102 (continued)
8) With the VIM baseboard’s component side (the side with the VIM
connectors) facing up, align the two VIM connectors on the Model 6231
VIM−2 module (J1 and J2) with two VIM connectors on the baseboard—
the Model 6231 VIM−2 module may be installed in either the top or
bottom mezzanine position on the VIM baseboard. See Figure 2−2 for
location of the Model 6231 VIM connectors, and see Figure 2−4 for location
of the VIM connectors on a typical VIM baseboard.
9) GENTLY but firmly, press down on the VIM module opposite the
connectors to fully seat the VIM module’s connectors into the baseboard’s.
If you meet with significant resistance, check the connector alignment.
NOTE:
Misalignment can cause bent pins or break connector
housings, so NEVER APPLY EXCESSIVE FORCE.
10) After seating the connectors, secure the front of the VIM module to the
VIM baseboard by screwing two short pan−head Phillips screws through
the holes at the front of the VIM module into the threaded holes in the
baseboard’s panel brackets (indicated at positions (a) and (b), or (c) and (d)
on Figure 2−4).
11) Turn the assembly over, such that the VIM module is on the work surface
and the solder side of the VIM baseboard is facing up. Secure the
baseboard to the nylon spacer on the VIM module, using the nylon screw
removed earlier, through the rear mounting hole on the baseboard
(indicated at position (e) or (f) on Figure 2−4).
12) Remove one top or bottom blank panel insert from the Option 102 front
panel assembly provided (see illustration below), depending on the VIM
position in which the Model 6231 module was installed in the above steps.
The remaining fixed plates and blank panel will be installed onto the
baseboard front panel in the following steps.
Top
fixed plate
Figure 2−11: Option 102 Front Panel Assembly
Rev.: B.4
Top blank panelBottom blank panel
Middle
fixed plate
Bottom
fixed plate
Page 33
Pentek Model 6230/6231 Operating ManualPage 33
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.2Installing Model 6231 with Option 102 (continued)
13) Install the Option 102 front panel assembly, which is in two pieces
depending on which blank panel you removed in the prior step, onto the
VIM baseboard front panel, using the four long pan−head Phillips screws
supplied (see following illustration).
Front Panel Bracket Screws
Option 102
Standoff
3
2
1
0
3
210
Figure 2−12: VIM Baseboard with Option 102 Front Panel Assembly
(Model 6231 illustrated in upper VIM position)
14) CAREFULLY, insert the two stacking connectors into the solder side of the
Model 6231 PCB. Note that the stacking connector’s longer leads must be
inserted through holes on the solder side of the PCB (see following
illustration).
Stacking Connectors
Stacking Connector
Long Leads
VIM PCB
(Side View)
Figure 2−13: Option 102 Stacking Connectors on VIM Module
Be careful when installing the stacking connectors, as they are fragile
CAUTION
and can easily be damaged.
Rev.: B.4
Page 34
Page 34Pentek Model 6230/6231 Operating Manual
2.3Installing the Model 6230/6231 on a VIM Baseboard (continued)
2.3.2Installing Model 6231 with Option 102 (continued)
15) Align the Option 102 PCB connectors J1 and J2 with the two stacking
connectors on the VIM module. GENTLY but firmly, press down on the
areas of the board opposite the connectors to fully seat the board’s
connectors into the VIM module.
NOTE:
Misalignment can cause bent pins or break connector
housings, so NEVER APPLY EXCESSIVE FORCE.
16) After seating the connectors, secure the front of the Option 102 PCB to the
baseboard by screwing two short pan−head Phillips screws through the
holes in the front of the PCB into the threaded holes in the Option 102
front panel assembly’s panel brackets (see illustration below) .
Standoff Screws (5)
Standoffs
Front Mounting Screws (2)
Figure 2−14: Option 102 PCB Mounted on VIM Module
Option 102 PCB
VIM PCB
VIM Panel Brackets
(Side View)
17) Screw the five pan−head Phillips standoff screws removed from the PCB
earlier through the solder side of the Option 102 PCB into the VIM module
standoffs (see illustration above).
18) Attach the Model 6231 and Option 102 front panels to the assembly, by
screwing the countersunk Phillips panel screws through the holes in each
VIM panel into the corresponding threaded holes in the front panel
brackets.
The Model 6231 installation is complete.
Rev.: B.4
Page 35
Pentek Model 6230/6231 Operating ManualPage 35
2.4Front Panel Connections
The Model 6230 has two (upper and lower) VIM front panels. The Model 6231 stan−
dard configuration has a single VIM front panel—a second front panel is provided with
Option 102, mounted on an adjacent board. The Model 6231 panels can be installed in
either the upper or the lower VIM position of the baseboard (Section 2.3). All panels
are illustrated in Figure 2−15, on the following page.
The front panel signal connectors are described in the following subsections. The front
panel LEDs are described in Section 2.5.
2.4.1Analog Input Connectors
The front panels provide threaded, coaxial SMA connectors for analog sig−
nal inputs, one for each A/D input channel. The Model 6230 has four analog
input connectors labeled CH1 IN to CH4 IN, two on each panel. The Model
6231 has two analog input connectors labeled CH1 IN and CH2 IN on its
standard panel.
The analog input signal to each connector must be in the range of 2 V
±1.0 V. Each input drives an Analog Devices OPA642 amplifier with
approximately 50
Ω input impedance, which amplifies the amplitude of the
signal by a factor of one.
2.4.2Sample Clock Input Connector
The Model 6230 upper panel, and Model 6231 standard panel, has a
threaded, coaxial SMA connector, labeled EXT CLK IN, for input of an
external clock. The external clock signal should be a TTL pulse train with a
45% to 55% duty cycle, or a sine signal of 1 to 3 V
This input clock can be used as the reference signal to derive the sample
clock signal for the A/D converters and digital receivers. Use bit D2, EXT
CLK, in the Master Control Register (see Section 3.7.8), to enable this input.
p−p
,
p−p
.
Rev.: B.4
Page 36
Page 36Pentek Model 6230/6231 Operating Manual
2.4Front Panel Connections (continued)
Upper
Panel
Lower
Panel
Model 6230
Figure 2−15: Models 6230 and 6231 Front Panels
Rev.: B.4
Standard
Panel
Model 6231
Option 102
Panel
Page 37
Pentek Model 6230/6231 Operating ManualPage 37
2.4Front Panel Connections (continued)
2.4.3Sync/Gate Connector
The 26−pin SYNC/GATE connector provides input/output pins for the
low−voltage differential signal (LVDS) Sync Bus: four gates, one sync, and
one clock. When the Model 6230/6231 is a sync bus Master, these pins out−
put the bus to other slave units. When the 6230/6231 is a Slave, these pins
input the signals from a Master. This connector also accepts two TTL gate
and sync inputs. The mating 26−pin connector is Pentek part # 353.02607
(ERNI part # 214346). The following table shows the connector pinouts.
Table 2−4: SYNC/GATE Connector Pins
SignalPinPinSignal
TTL GATEB1A1GND
TTL SYNCB2A2GND
GATE D−B3A3GATE D+
GNDB4A4GND
GATE C−B5A5GATE C+
GNDB6A6GND
GATE B−B7A7GATE B+
GNDB8A8GND
GATE A−B9A9GATE A+
GNDB10A10GND
SYNC−B11A11SYNC+
GNDB12A12GND
CLK−B13A13CLK+
B1
B2
B3
B4
B5
B6
B7
B8
B9
B10
B11
B12
B13
A1
A2
A3
A4
A5
A6
A7
A8
A9
A10
A11
A12
A13
2.4.4Sync/Gate Header
The 4−pin SYNC/GATE/TRG connector [optional on the Model 6231] pro−
vides a second set of pins for TTL sync/gate inputs. These are wired to the
same TTL SYNC and GATE signals as pins B2 and B1 of the SYNC/GATE
connector, above. The mating 2−pin connector housing (use two for both
signals) is Pentek part # 353.00201 (Berg # 65039−035) or equivalent. This
housing uses discrete 0.025" square socket pins, Pentek part # 354.00104
(Berg # 48254−000). The following table shows the connector pinouts.
TTL SYNC12GND
TTL GATE/TRG34GND
Table 2−5: SYNC/GATE/TRG Header Pins
SignalPinPinSignal
1
3
2
4
Rev.: C.1
Page 38
Page 38Pentek Model 6230/6231 Operating Manual
2.4Front Panel Connections (continued)
2.4.5FPGA Connector
The 50−pin FPGA connector for the Model 6230 includes two 16−pin FPGA
input/output data paths. The pins correspond to bits D15 through D0 of the
I/O Data Register for each of the FPGAs (see Section 3.22). This 50−pin
connector is optional on Model 6231 (Option 102), and includes one 24−pin
I/O data path. These pins correspond to bits D15 through D0 of the I/O
Data Register (see Section 3.22), and bits D16 through D23 of the I/O Data 2
Register (see Section 3.22). The mating connector is Pentek part # 353.05006
(ERNI part # 214347). The following table shows the connector pin configu−
ration for both VIM models.
The Model 6230 front panel has eight LED indicators [six LEDs on the Model 6231], as
illustrated in Figure 2−15, on page 36.
2.5.1Clock (CLK) LED
The green CLK LED is illuminated when a valid clock signal is detected. If
this LED is not illuminated, then no clock has been detected and no data
from the input stream can be processed.
2.5.2Over Temperature (TEMP) LED
There are four temperature/voltage sensors on the Model 6230 PCB [two
sensors on the Model 6231 PCB]. The temperature sensor thresholds are set
by a baseboard processor (see Hardware Monitor Port Register, Section 3.6).
When an over−temperature (or over−voltage) condition is indicated, the red
TEMP LED is illuminated on the front panel. In addition, two TEMP inter−
rupts from these sensors are available to all processors on the VIM base−
board (see Table 3−18, Section 3.14).
NOTE:
You must set up the temperature/voltage sensors’ Hardware
Monitor Port following power on of the VIM baseboard.
ReadyFlow board support software is provided for this purpose
(see Hardware Monitor Port Register, Section 3.6).
2.5.3Master (MAS) LED
The yellow MAS LED illuminates when this Model 6230/6231 is the sync
bus Master (MASTR bit D00 = 1, Master Control Register, Section 3.7.10).
The bus Master generates all sync/gate/clock signals on the sync bus.
2.5.4Terminate (TRM) LED
When this Model 6230/6231 is the last (or only) Slave unit on the sync bus,
you must enable bus termination (TERM bit D01, Master Control Register,
Section 3.7.9). Enabling bus termination illuminates the yellow TRM LED.
2.5.5Overload (OVLD CHn) LEDs
There are four red overload LEDs on the Model 6230 [two on the Model
6231], one for each A/D input, labeled OVLD CH1 to OVLD CH4 [OVLD CH1 to OVLD CH2]. Each LED is an indicator for an A/D overload detec−
tion function in each of the AD6644 A/D converters. When an overload
indication is set by the AD6644, the associated OVLD LED is illuminated. In
addition, an OVLD interrupt may be generated from each A/D overload
indication to all VIM baseboard processors (see Table 3−18, Section 3.14).
Rev.: B.4
Page 40
Page 40Pentek Model 6230/6231 Operating Manual
This page is intentionally blank
Rev.: B.4
Page 41
Pentek Model 6230/6231 Operating ManualPage 41
Chapter 3: Memory Maps and Register Descriptions
3.1Overview
This chapter describes processor access to the Model 6230 and Model 6231 from the
VIM baseboard. Memory maps to VIM module resources are given from the baseboard
processor’s viewpoint, and details are provided describing the use of each resource.
3.2Model 6230/6231 Memory Map
The two tables below provide base addresses of the control and status registers for each
type of Pentek VIM Baseboard. Use the applicable base address from these tables as the
VIM registers base address (‘
For a Pentek Model 4290 through 4295 VIM baseboard, use the VIM base addresses in
Table 3−1, below. On these baseboards, each node processor uses the same VIM base
address, which refers to the VIM Node Interface connected to that processor.
VIM_base’) in the following sections.
Table 3−1: VIM Base
Addresses for Models 4290 to 4295 VIM Baseboards
VIM BaseboardVIM Registers Base Address (VIM_base)
Model 4290 or 42910x0032 0000
Model 4292 or 42930x0202 0000
Model 4294 or 42950x1D02 0000
For a Pentek Model 4205 VIM baseboard, use the VIM base addresses in Table 3−2,
below. The Model 4205 has a single baseboard processor, which uses a separate base
address for each VIM Node Interface (emulating four processors). In addition, on this
baseboard the VIM module registers can be accessed from the VMEbus.
Table 3−2: VIM Base Addresses for Model 4205 VIM Baseboards
VIM Registers Base Address (VIM_base)
VIM
MPC7455VMEbus A32 Slave *
VIM A0x4082 0000A32VME_base + 0x00B2 0000
VIM B0x4086 0000A32VME_base + 0x00B6 0000
VIM C0x408A 0000A32VME_base + 0x00BA 0000
VIM D0x408E 0000A32VME_base + 0x00BE 0000
* Offsets relative to the base address ‘A32VME_base’ set on the Model 4205 baseboard (refer
to the Model 4205 Operating Manual for these settings)
Rev.: B.4
Page 42
Page 42Pentek Model 6230/6231 Operating Manual
3.2Model 6230/6231 Memory Map (continued)
The following table provides a memory map for accessing the Model 6230/6231 control
registers. All register addresses are expressed as offsets from the VIM base address for
th e applicable VIM baseboard—see Table 3−1 or Table 3−2 on the prior page for the VIM
registers base address (‘VIM_base’) for your VIM baseboard. The subsections follow−
ing the table provide detailed information about each register.
* VIM_base = see Table 3−1 or Table 3−2 on prior page
Rev.: B.4
Page 43
Pentek Model 6230/6231 Operating ManualPage 43
3.3Virtex Config Register
The Virtex Config Register is used to reconfigure/reprogram the Virtex FPGAs. The
bits in this register allow you to read and set the status of the FPGA configuration cycle,
and to upload the configuration from a VIM processor. Only Processor A and Proces−
sor C on the VIM baseboard have their own version of this register, for the Virtex FPGA
associated with that processor (for example, FPGA1 for Processor A).
Refer to Section 1.10, FPGA Configuration, for additional information about configur−
ing the gate arrays.
The following table shows the contents of the Virtex Config Register. The subsections
following the table provide descriptions of the bits in this register.
Table 3−4: Virtex Config Register
R/W @ VIM_base+0x0004 (Proc A & C only)
D15 – D06D05D04*D03*D02*D01D00
Bit Name
Function
Not usedWRITEBUSYINITDONEPRGMLD SRC
Write zeros,
Mask read
All bits default to the logic '0' state at power on and reset
0 = Disable
1 = Write
0 = Disable
1 = Enable
* These bits are Read Only
0 = Configuring
1 = Completed
0 = Configuring
1 = Completed
0 = Disable
1 = Reprogram
0 = Onboard
1 = Upload
3.3.1WRITEBit D05
This bit sets the write access of the Virtex Config Data Register, described in
Section 3.4. When you set this bit to logic '1', you can write configuration
data to the FPGA. To write data with the Virtex Config Data Register, you
must also enable configuration Processor Upload (LD SRC bit D00 = 1, next
page). Clear the bit to logic '0' (its default state) to disable configuration data
from being accepted by the FPGA.
3.3.2BUSYBit D04
This read−only bit indicates the status of the Virtex FPGA’s ‘BUSY’ pin. It is
only meaningful in processor upload mode, though not needed. (For refer−
ence only.)
3.3.3INIT Bit D03
This read−only bit indicates the status of the Virtex FPGA’s ‘INIT
initialization of a configuration cycle this bit goes to logic '0', then logic '1'. It
will remain at logic '0' if an error in configuration data is detected. When
read as logic '1', initialization is done.
’ pin. At
Rev.: B.4
Page 44
Page 44Pentek Model 6230/6231 Operating Manual
3.3Virtex Config Register (continued)
3.3.4DONEBit D02
This read−only bit indicates the status of the Virtex FPGA’s ‘DONE’ pin.
When read as logic '0', the FPGA is in the configuration cycle. When read as
logic '1', the FPGA has completed configuration.
3.3.5PRGMBit D01
The FPGA begins a configuration reprogramming cycle after you transition
this bit from logic '0', to logic '1', then to logic '0', while in Serial PROM or
Processor Upload mode (see LD SRC, below). It has no effect in the Serial
Download mode, as the X−Checker cable has control in that mode.
3.3.6LD SRCBit D00
This bit selects the source of the Virtex FPGA configuration data. When you
clear this bit to logic '0' (its default state), the FPGA selects its configuration
data from the on−board Serial PROM or from a Serial Download, depending
on the setting of the configuration data source jumper on the module’s PCB,
see Section 2.2. When you set this bit to logic '1', the FPGA is in Processor
Upload mode, and you can upload the configuration from the processor
using the Virtex Config Data Register, as described in Section 3.4 below.
3.4Virtex Config Data Register
The Virtex Config Data Register writes configuration data to the Virtex FPGA. Write
access is set using the WRITE bit in the Virtex Config Register, Section 3.3.1. When LD
SRC = 1, Section 3.3.6, FPGA configuration data can be written, one byte at a time,
using this register. Processors A and C on the VIM baseboard have their own version
of this register, for the Virtex FPGA associated with that processor (for example,
FPGA1 for Processor A). The following table shows the contents of this register.
Table 3−5: Virtex Config Data Register
W.O. @ VIM_base+0x0008 (Proc A & C only)
D15 – D08D07D06D05D04D03D02D01D00
Bit Name
Function
Not usedD7D6D5D4D3D2D1D0
Write zeros,
Mask read
When reset or power−up, the state of this register is unknown.
Eight bits (one byte) of configuration data
Refer to Section 1.10, FPGA Configuration, for additional information about configur−
ing the Virtex FPGA.
Rev.: B.4
Page 45
Pentek Model 6230/6231 Operating ManualPage 45
3.5Wait States Register
With the possibility of DBCLKs on future processor boards above 50 MHz, program−
mable wait states are provided to guarantee adequate access time. These wait states are
user−programmable, using the Wait States Register. Each processor on the VIM base−
board has its own version of this register for the devices associated with that processor
(for example, Processor A can set the wait states for FPGA1 and CG4016 A & B).
The following table shows the contents of this register. The paragraphs following the
table provide descriptions of these bits.
Number of wait states for DDRsNumber of wait states for FPGAs
All bits default to the logic '1' state at power on and reset
NOTE:
At power up, the wait states default to a maximum.
The binary value in the four GC DDR bits determines the number of wait states for both
GC4016 DDRs (wait state = [GC DDR] + 2 DBCLK cycles).
The binary value in the four CTL REG bits sets the number of wait states for the FPGA
(wait state = [CTL REG] + 2 DBCLK cycles).
The wait states are equal to the greater of the VIM interface minimum write cycle time
or the wait state selection in this register.
For Model 6230/6231 boards with a DBCLK of 50 MHz or less, a setting of GC DDR =
0x2 and CTL REG = 0x2 is adequate.
Rev.: B.4
Page 46
Page 46Pentek Model 6230/6231 Operating Manual
3.6Hardware Monitor Port Register
The board’s Voltage/Temperature Monitors, Analog Devices ADM1024s, provide con−
stant monitoring of critical voltages and temperatures on the PCB. There are four tem−
perature sensors on the Model 6230 [two on the Model 6231], which are controlled by
the Hardware Monitor Port. When an over−temperature or over−voltage condition is
indicated, the red TEMP LED is illuminated on the front panel. In addition, an over−
temperature/voltage interrupt from each ADM1024 is available to all VIM baseboard
processors (see Table 3−18, Section 3.14).
There are several programmable temperature and voltage sensors controlled by these
ADM1024s. The Hardware Monitor Port Register allows you to configure the sensor
thresholds. The Hardware Monitor Port Register is accessible only to Processor A on
the VIM baseboard (or to Processor C for a Model 6231 that is installed in the bottom
VIM position on the baseboard).
The following table shows the contents of this register. The register contains three bits:
one bit sets the read/write direction of the register and the other two provide the serial
clock and data to/from the Hardware Monitor Port.
Bit Name
Function
Note
Table 3−7: Hardware Monitor Port Register
R/W @ VIM_base+0x0010 (Proc A or C only)
D15 – D03D02D01D00
Not usedSER DIRSER CLKSER DAT
Write with zeros, Mask when reading
All bits default to the logic '0' state at power on and reset
0 = Write
1 = Read
Serial
Clock
Serial
Data
You must set up the Hardware Monitor Port and the port devices following
power on of the VIM baseboard.
Routines for setup and programming the temperature and voltage thresholds are pro−
vided in the ReadyFlow board support software for the Model 6230 (Pentek part
#801.62300).
Alternately, you can program the voltage and temperature thresholds directly using
the Hardware Monitor Port Register. The instructions for programming the ADM1024
with this register are provided on the following page.
Rev.: B.4
Page 47
Pentek Model 6230/6231 Operating ManualPage 47
3.6Hardware Monitor Port Register (continued)
The Model 6230 uses two ADM1024’s. Each ADM1024 controls several temperature
sensors and voltage sensors on the PCB. Both ADM1024’s can be accessed only from
processor A using the Hardware Monitor Port Register. The Model 6231 uses one
ADM1024, which can be accessed only from processor A or C, depending on its
installed VIM position, using the Hardware Monitor Port Register.
You can set their limits using internal ADM1024 registers (refer to the ADM1024 data
sheet, Appendix D, for description of these registers). These ADM1024 registers are
programmed from the Hardware Monitor Port Register and are addressed using sepa−
rate bus read and write addresses, sent as serial data through the port.
•On the Model 6230 with two ADM1024’s, the bus addresses are ("Upper" and
“Lower” refer to the associated VIM position on the baseboard for each ADM1024):
Upper ADM1024 − bus write address: 58 hex
Upper ADM1024 − bus read address: 59 hex
Lower ADM1024 − bus write address: 5A hex
Lower ADM1024 − bus read address: 5B hex
•On the Model 6231 with one ADM1024, the bus addresses are:
Bus write address: 58 hex
Bus read address: 59 hex
There are eight conditions on the PCB monitored by each ADM1024. These conditions,
the registers used, and their limits are identified in the following table:
Table 3−8: ADM1024 Registers
Monitored ConditionADM1024 RegisterLimits
Internal temperatureInternal Temp Value0 to 85 degrees C
External temperature sensor 1 (under shield)Ext. Temp1 Value0 to 85 degrees C
External temperature sensor 2 (on PCB)+2.5V Measured Value/EXT Temp20 to 85 degrees C
+1.8 Volt supply+Vccp1 Measured Value+1.8 Volts, ±10%
+2.5 Volt supplyFAN2/AIN2 Value+2.5 Volts, ±10%
+3.3 Volt supplyVcc Measured Value+3.3 Volts, ±10%
+5 Volt supply+5V Value+5 Volts, ±10%
+12 Volts Supply+12V Measured Value+12 Volts, ±10%
Refer to the ADM1024 data sheet, Appendix D, for more information on reading and
writing from/to this device.
Rev.: B.4
Page 48
Page 48Pentek Model 6230/6231 Operating Manual
3.7Master Control Register
The Master Control Register allows you to configure the Model 6230 or Model 6231 as a
Master or Slave on the LVDS sync bus, toggle the on−board sync bus termination, select
the source of the clock, select a clock divider, select the source and polarity of the sync,
and enable the on−board oscillator. The Master Control Register is accessible only to
Processor A on the VIM baseboard (or Processor C for a Model 6231 that is installed in
the bottom VIM position on the baseboard).
The following table shows the contents of the Master Control Register. The subsections
following the table provide descriptions of the bits and fields in the register.
Table 3−9: Master Control Register
R/W @ VIM_base+0x0020 (Proc A or C only)
D15D14 – D10D09D08
Bit Name
Function
Bit Name
Function
RESETNot usedSYNC SRC SYNC POL
0 = Run
1 = Reset
Write with zeros,
Mask when reading
0 = Bypass
1 = Sync Bus
0 = Negative
1 = Positive
D07D06D05D04D03D02D01D00
EXT SYNC
EN
0 = Register
1 = Ext Sync
All bits default to the logic '0' state at power on and reset
MCLK DIV1 MCLK DIV0
00 = Divide by 1
01 = Divide by 2
10 or 11 = Divide by 4
CLK SRC
SEL
0 = Bypass
1 = Sync Bus
OSC DSBLEXT CLKTERMMASTR
0 = Enable
1 = Disable
0 = Oscillator
1 = Ext Clock
0 = None
1 =
Terminat’d
0 = Slave
1 = Master
3.7.1RESETBit D15
This bit issues a reset to the entire Model 6230/6231 VIM module. When
this bit is set to logic '1', the board is in reset. When the bit is cleared to logic
'0' (its default state), the board is in a normal run state.
The Reset signal should be held for at least four times the front panel
Note
clock speed (4 x Front Panel Clock).
3.7.2SYNC SRCBit D09
This bit selects the source of the sync signal for the GC4016 DDRs from
either the LVDS sync bus or an on−board sync. When this bit is cleared to
logic '0' (its default state), the sync bus is bypassed and the sync is selected
by EXT SYN EN bit D07, on the next page. When the bit is set to logic '1', the
sync bus (SYNC/GATE connector, Section 2.4.3) is the source of the sync.
Rev.: B.4
Page 49
Pentek Model 6230/6231 Operating ManualPage 49
3.7Master Control Register (continued)
3.7.3SYNC POLBit D08
This bit selects the polarity of the external sync signal for resetting the
GC4104 DDRs. When this bit is cleared to logic '0' (its default state), a nega−
tive sync resets the GC4016s. When the bit is set to logic '1', a positive sync
resets the GC4016s.
3.7.4EXT SYNC ENBit D07
When this Model 6230/6231 is the sync bus Master (MASTR bit D00 = 1) or
when the LVDS sync bus sync input is bypassed (SYNC SRC bit D09 = 0),
this bit selects the sync signal for the GC4016 DDRs from either an on−board
register or the external TTL sync input. When this bit is cleared to logic '0'
(its default state), the register (Sync/Gate Generator Register, Section 3.13) is
the source of the sync. When the bit is set to logic '1', the external TTL sync
input (SYNC/GATE connectors, Section 2.4.3 or 2.4.4) is the source.
3.7.5MCLK DIVnBits D06 to D05
These two bits select the master clock divider when this Model 6230/6231 is
a sync bus Master (MASTR bit D00 = 1). This clock divider applies to the
clock selected by EXT CLK bit D02, below. The settings for these bits are:
'00' (default)- divide by 1
'01'- divide by 2
'10' or '11'- divide by 4
3.7.6CLK SRC SELBit D04
This bit selects the source of the module’s clock from either the LVDS sync
bus or an on−board clock source. When the bit is cleared to logic '0' (its
default state), the sync bus is bypassed and the clock is selected by EXT CLK
bit D02, below. When this bit is set to logic '1', the sync bus (SYNC/GATE
connector, Section 2.4.3) is the source of the clock.
3.7.7OSC DSBLBit D03
This bit enables or disables the output from the on−board oscillator. When
the bit is cleared to logic '0' (its default state), the on−board oscillator is
enabled. When this bit is set to logic '1', the on−board oscillator is disabled.
Rev.: B.4
Page 50
Page 50Pentek Model 6230/6231 Operating Manual
3.7Master Control Register (continued)
3.7.8EXT CLKBit D02
When this Model 6230/6231 is a sync bus Master (MASTR bit D00 = 1) or
when the sync bus clock is bypassed (CLK SRC bit D04 = 0), this bit selects
the module’s clock signal from either the on−board oscillator or the external
clock input. When this bit is cleared to logic '0' (its default state), the on−
board oscillator is selected (for this setting, OSC DSBL bit D03, above, must
be cleared to logic '0' to enable the oscillator). When the bit is set to logic '1',
the external clock input is used (EXT CLK IN connector, Section 2.4.2).
3.7.9TERMBit D01
This bit enables termination of the sync bus by a Slave unit, or a Master unit
if it is the only unit on the bus. When the bit is cleared to logic '0' (its default
state), the sync bus is not terminated. When the bit is set to logic '1', the bus
is terminated. The sync bus must be terminated when this Model 6230/6231
is the last (or only) Slave on the sync bus, or if it is a Master and the only unit
on the bus.
3.7.10MASTRBit D00
This bit selects either Master or Slave on the sync bus for this Model 6230/
6231. When the bit is cleared to logic '0' (its default state), the module is a
Slave. When bit is set to logic '1', the sync bus on this board is Master.
When set to a sync bus Master, this Model 6230/6231 is the source of the
sync, clock, and gate signals output to the LVDS sync bus. You must select
these sources using control bits EXT SYNC EN (bit D07), EXT CLK (bit D02),
and EXT GATE (Section 3.10.9), and if needed, generate the sync and/or
gate signals using the Sync/Gate Generator Register (Section 3.13).
Rev.: B.4
Page 51
Pentek Model 6230/6231 Operating ManualPage 51
3.8Bypass Rate Divide Register
This register sets the decimation rate of data samples written to the processor BIFO
when the module is in DDR Bypass mode (see Section 4.2, Data Routing and Formats).
Each processor on the VIM baseboard has its own version of this register for the BIFO
associated with that processor.
The following table shows the contents of this register. The paragraph following the
table provide descriptions of these bits.
Table 3−10: Bypass Rate Divide Register
R/W @ VIM_base+0x0024
D15 – D12D11D10D09D08
Bit Name
Function
Bit Name
Function
Note
Not usedB11B10B09B08
Write with zeros, Mask when reading4 MSBs of 12−bit rate divider 'N−1'
D07D06D05D04D03D02D01D00
B7B6B5B4B3B2B1B0
8 LSBs of 12−bit rate divider 'N−1'
All bits default to the logic '0' state at power on and reset
The rate divider is a 12−bit binary value, with bit D11 the MSB. To take every Nth
sample, where N is any integer value between 1 and 4096, set this register to N−1. The
default value is '0x000', which is dividing by 1.
If you select ‘Channel Packing’ DDR Bypass mode (Section 4.2.3), the bypass
rate dividers of both processor channels (A and B, or C and D) MUST be set at
the same rate and synchronized for proper operation.
Rev.: B.4
Page 52
Page 52Pentek Model 6230/6231 Operating Manual
3.9Channel Enable Register
The Channel Enable Register enables the output from any of the eight receiver channels
of the two GC4016s associated with a VIM baseboard processor, to a processor’s BIFO.
It has no effect in DDR Bypass mode (see Section 4.2). Each processor on the VIM base−
board has its own version of this register for the BIFO associated with that processor.
The following table shows the contents of this register. The paragraph following the
table provide descriptions of these bits.
Table 3−11: Channel Enable Register
R/W @ VIM_base+0x0028
D15 – D08D07D06D05D04D03D02D01D00
Bit Name
Function
Not usedEN7EN6EN5EN4EN3EN2EN1EN0
Write zeros,
Mask read
All bits default to the logic '0' state at power on and reset
0 = Disable output
1 = Enable output
Each register bit corresponds to one of the GC4016 channels for this processor. When a
bit is cleared to logic '0', output from the associated GC4016 channel is disabled. When
a bit is set to logic '1', the output is enabled to the associated BIFO for this channel. The
following table gives the DSP/GC4016 channel assignments for each bit.
Table 3−12: Register Bit Channel Assignments
Register BitDSP ChannelGraychipGC4016 Channel
EN0DSP CH1
EN1DSP CH2Channel B
EN2DSP CH3Channel C
EN3DSP CH4Channel D
EN4DSP CH5
EN5DSP CH6Channel B
EN6DSP CH7Channel C
EN7DSP CH8Channel D
First GC4016
Second GC4016
Channel A
Channel A
When you enable any DDR channel outputs, ensure that the Graychip GC4016s
Note
Rev.: B.4
are configured accordingly, by writing control information into control registers
in the GC4016s (see Section 3.23 for the Graychip Registers).
Page 53
Pentek Model 6230/6231 Operating ManualPage 53
3.10Gate Control Register
The Gate Control Register contains several fields that select the characteristics and
source of the BIFO write gates for the controlling processor. Each processor on the VIM
baseboard has its own version of this register. However, only Processors A and C can
set the gate polarity (GATE POL, bit D08). Each processor may use any one of the gates
available (Gate A, B, C, or D) for enabling BIFO writes associated with that processor.
When the Model 6230/6231 is the sync bus Master (MASTR = 1, Master Control Regis−
ter, Section 3.7.10), the board generates four gates for the LVDS sync bus: Gates A and C
from Processor A (or Processor C for a Model 6231 that is installed in the bottom VIM
position on the baseboard), and Gates B and D from Processor B (or Processor D).
When the sync bus is bypassed (GATE SRC bit D04 = 0), the Model 6230/6231 gener−
ates only Gate A and B for selection on the board. Refer to Chapter 5, Gate/Sync/Clock
Description, for additional information about these gate and sync signals.
The following table shows the contents of the Gate Control Register. The subsections
following the table give descriptions of the bits in this register.
Bit Name
Function
Bit Name
Function
Table 3−13: Gate Control Register
R/W @ VIM_base+0x0030
D15 – D13D12D11D10D09D08
Not usedINT EDGE D INT EDGE C INT EDGE B INT EDGE A GATE POL
Write with zeros,
Mask when reading
0 = Start
1 = End
0 = Start
1 = End
0 = Start
1 = End
0 = Start
1 = End
0 = Negative
1 = Positive
D07D06D05D04D03D02D01D00
GATE SEL1 GATE SEL0 GATE DISBL GATE SRC
00 = Gate A
01 = Gate B
10 = Gate C
11 = Gate D
All bits default to the logic '0' state at power on and reset
0 = Enable
1 = Disable
0 = External
1 = Sync
Bus
TRIG
CLEAR
0 = None
1 = Force
Gate inactive
HOLD
MODE
0 = None
1 = Hold
GATE/TRIG EXT GATE
0 = Gate
1 = Trigger
0 = Register
1 = TTL
Input
3.10.1INT EDGE xBit D12/D11/D10/D09
Each of these four bits selects whether the start or the end of the respective
gate (Gate A for INT EDGE A, etc.) creates an interrupt to the controlling
processor (each gate interrupt may be independently enabled for this pro−
cessor using the Interrupt Mask Register, Section 3.14). When the bit is
cleared to logic '0' (its default state), the start of the gate creates an interrupt.
When the bit is set to logic '1', the end of the gate creates the interrupt.
Rev.: B.4
Page 54
Page 54Pentek Model 6230/6231 Operating Manual
3.10Gate Control Register (continued)
3.10.2GATE POLBit D08
This bit selects the polarity of the external TTL gate/trigger input. This bit is
only available to Processors A and C (Processor A controls polarity for B;
Processor C controls polarity for D). When the bit is cleared to logic '0' (its
default state), BIFO writes are enabled by a negative input in Gate mode, or
triggered on a negative−going edge in Trigger mode. When the bit is set to
logic '1', BIFO writes are enabled by a positive input in Gate mode, or trig−
gered on a positive−going edge in Trigger mode.
3.10.3GATE SELnBits D07 to D06
These two bits select the gate for the BIFO writes associated with this pro−
cessor on the VIM baseboard. The settings for these two bits are:
'00' (default)- Gate A
'01'- Gate B
'10'- Gate C
'11'- Gate D
3.10.4GATE DISBLBit D05
This bit enables or disables the selected gate (GATE SEL bits D07 to D06,
above) from controlling BIFO writes for this processor. When this bit is
cleared to logic '0' (its default state), the gate is enabled to control BIFO
writes. When this bit is set to logic '1', the gate is disabled. When the gate is
disabled, BIFO writes are defaulted to enabled; however, it does not disable
the generation of interrupts from the gate signal.
3.10.5GATE SRCBit D04
This bit selects the source of the selected BIFO gate (GATE SEL bits D07 to
D06, above) from either the LVDS sync bus or an external/on−board source.
When this bit is cleared to logic '0' (its default state), the sync bus is
bypassed and the BIFO gate is selected using EXT GATE bit D00, next page.
When the bit is set to logic '1', the LVDS sync bus (SYNC/GATE connector,
Section 2.4.3) is the source of the gate.
3.10.6TRIG CLEARBit D03
This bit forces the selected gate to the inactive state in Trigger mode (GATE/
TRIG bit D01 = 1, next page). When this bit is cleared to logic '0' (its default
state), there is no effect on the gate. When the bit is set to logic '1', the gate is
forced to inactive (BIFO writes disabled), regardless of the trigger length
specified (Trigger Length Register, Section 3.11).
Rev.: B.4
Page 55
Pentek Model 6230/6231 Operating ManualPage 55
3.10Gate Control Register (continued)
3.10.7HOLD MODEBit D02
This bit enables a gate Hold after the trigger is received in Trigger mode
(GATE/TRIG bit D01 = 1, below). When the bit is cleared to logic '0' (its
default state), the selected gate is active (BIFO writes enabled) for the speci−
fied trigger length after the trigger (specified using the Trigger Length Reg−
ister, Section 3.11), and then goes inactive (BIFO writes disabled). When the
bit is set to logic '1', Hold is enabled and the gate remains active (BIFO
writes enabled) after the trigger is received until this bit is cleared.
3.10.8GATE/TRIGBit D01
When the source of the selected BIFO gate is either an on−board register or
the external TTL gate input (EXT GATE bit D00, below), this bit selects
between Gate or Trigger mode for enabling BIFO writes. When the bit is
cleared to logic '0' (its default state), Gate mode is selected. When the bit is
set to logic '1', Trigger mode is selected, and you must set the trigger length
using the Trigger Length Register, Section 3.11, or enable trigger Hold
(HOLD MODE bit D02, above).
3.10.9EXT GATEBit D00
When the Model 6230/6231 is the sync bus Master (MASTR = 1, Master
Control Register, Section 3.7.10), or when LVDS sync bus is bypassed for gate
signals (GATE SRC bit D04 = 0), this bit selects the source of the designated
BIFO gate either from an on−board register generator or from the external
TTL gate input. When this bit is cleared to logic '0' (its default state), the on−
board register (Sync/Gate Generator Register, Section 3.13) is the source of
the BIFO gate. When the bit is set to logic '1', the external TTL gate input
(SYNC/GATE connectors, Section 2.4.3 or 2.4.4) is the source of the gate.
See Chapter 5, Gate/Sync/Clock Description, for additional information about the use
and programming of gate and sync signals.
Rev.: B.4
Page 56
Page 56Pentek Model 6230/6231 Operating Manual
3.11Trigger Length Register
When Trigger mode is selected for a BIFO gate (GATE/TRIG = 1, Gate Control Regis−
ter, Section 3.10.8), this register sets the length that the gate is active (BIFO writes
enabled) after receipt of the trigger. Each processor on the VIM baseboard has its own
version of this register for the gate selected by that processor.
The following table shows the contents of this register. The paragraph following the
table provide descriptions of these bits.
Table 3−14: Trigger Length Register
R/W @ VIM_base+0x0034
D15 – D14D13D12D11D10D09D08
Bit Name
Function
Bit Name
Function
Note
Not usedD13D12D11D10D9D8
Write with zeros,
Mask when reading
6 MSBs of 14−bit Trigger Length
D07D06D05D04D03D02D01D00
D7D6D5D4D3D2D1D0
8 LSBs of 14−bit Trigger Length
All bits default to the logic '0' state at power on and reset
The trigger length is a 14−bit binary value, with bit D13 the MSB. This value specifies
the length of the gate after the trigger as the number of BIFO writes of the processor
that generates the selected gate, up to 16,383.
If you set the Trigger Length Register and select this gate source for multiple
processor channels (GATE SEL, Gate Control Register, Section 3.10.3), you
must set the same data routing modes for those channels (DAT MODE,
Channel Control Register, Section 3.12.4)
When in Master/Slave Mode only, there are several additional clock cycles
between when the Master Gate is turned off and when the BIFO write logic is
finally disabled. During that time period, depending on the Bypass Rate Divider
setting (Section 3.8), several additional samples will be written into the VIM
BIFO. This is unavoidable, because of the register delays in both the SYNC
Note
Rev.: B.4
bus Gate transmit and SYNC bus Gate receive logic. For any given setting of
the Bypass Rate Divider, however, the number of extra clock cycles is
absolutely deterministic and repeatable. One can, therefore, compensate for
the extra clocks incurred due to SYNC bus delays by reducing the Trigger
Length setting accordingly.
Page 57
Pentek Model 6230/6231 Operating ManualPage 57
3.12Channel Control Register
The Channel Control Register selects the data routing mode of each processor channel,
and contains bits to reset the data formatter state machine (in the associated FPGA), to
enable the sync signal to reset the DDR Bypass mode decimate dividers, and to reset
the processor channel. Each processor has its own version of this register, associated
with that processor channel on the VIM baseboard.
The following table shows the contents of the Channel Control Register. The subsec−
tions following the table provide descriptions of the bits in this register.
Table 3−15: Channel Control Register
R/W @ VIM_base+0x0038
D15D14 − D05D04D03D02D01D00
Bit Name
Function
RESETNot usedDIV RST ENFMTR EN DAT MODE2 DAT MODE1 DAT MODE0
0 = Run
1 = Reset
All bits default to the logic '0' state at power on and reset
Write with zeros,
Mask when reading
0 = Disable
1 = Enable
0 = Disable
1 = Enable
Data Routing mode −
see Section 3.12.4
3.12.1RESETBit D15
This bit issues a reset to the associated processor channel (A, B, C, or D).
(This does not reset the Master Control Register, Section 3.7). When the bit is
set to logic '1', the processor channel is in reset. When this bit is cleared to
logic '0' (its default state), the channel is in a normal run state.
3.12.2DIV RST ENBit D04
This bit enables the sync signal to reset the DDR Bypass mode rate dividers,
to allow synchronization of all dividers. When this bit is cleared to logic '0'
(its default state), this reset is not enabled. When bit is set to logic '1', the
sync is enabled to reset the bypass mode rate dividers.
3.12.3FMTR ENBit D03
This bit enables or resets the data formatter state machine (in the associated
FPGA). When this bit is cleared to logic '0' (its default state), the formatter is
disabled and reset. When the bit is set to logic '1', the formatter is enabled
for normal operation.
Rev.: B.4
Page 58
Page 58Pentek Model 6230/6231 Operating Manual
3.12Channel Control Register (continued)
3.12.4DAT MODEnBits D02 to D00
These three bits select the data routing mode from the two GC4016 DDRs to
the FPGA associated with this processor channel (for example, FPGA1 for
Processor A). There are seven different modes of data formatting (these
modes are described in Section 4.2). The settings for these three bits are:
See Section 4.2, Data Routing and Formats, for additional information about
the data formatting modes and operation of the FPGAs.
Note
When you select any data routing mode, ensure that the Graychip GC4016s
are configured accordingly, by writing control information into control
registers in the GC4016s (see Section 3.23 for the Graychip Registers).
The Channel Control Registers do not set up the GC4016 registers.
Rev.: B.4
Page 59
Pentek Model 6230/6231 Operating ManualPage 59
3.13Sync/Gate Generator Register
The Sync/Gate Generator Register is used on a Model 6230/6231 that is configured as a
sync bus Master (MASTR = 1, Master Control Register, Section 3.7.10), or on a board that
is not connected to an external LVDS sync bus. Each processor on the VIM baseboard
has its own copy of this register, associated with that processor channel.
The following table shows the contents of the Sync/Gate Generator Register. The reg−
ister contains two bits that generate gate and sync signals for the associated processor
channel. The subsections following the table give descriptions of these bits.
Table 3−16: Sync/Gate Generator Register
R/W @ VIM_base+0x003C
D15 − D02D01D00
Bit Name
Function
Not usedFIFO GATESYNC
Write with zeros,
Mask when reading
All bits default to the logic '1' state at power on and reset
0 = Enable
1 = Disable
0 = Reset
1 = Release
3.13.1FIFO GATE Bit D01
If the gate source is set to the on−board register (EXT GATE = 0, Gate Con−
trol Register, Section 3.10.9), this bit creates the gate signal (GATE A, B, C, or
D) for the associated processor BIFO, and for output to the sync bus (SYNC/
GATE connector, Section 2.4.3) for a sync bus Master. When the bit is cleared
to logic '0', BIFO writes are enabled. When the bit is set to logic '1' (its
default state), BIFO writes are disabled.
3.13.2SYNC Bit D00
If the sync source is set to the on−board register (EXT SYNC EN = 0, Master
Control Register, Section 3.7.4), this bit creates the sync signal (SYNC) for all
GC4016s, and for output to the sync bus for a sync bus Master (SYNC/
GATE connector, Section 2.4.3). The SYNC
together to create the SYNC signal for the board. When the bit is cleared to
logic '0', the GC4016 is in reset. When bit is set to logic '1' (its default state),
the GC4016 is released for normal operation.
bits for all processors are Or’ed
Set the SYNC bit to logic '1' for all processors that are not used to create the
Note
sync signal.
Refer to Chapter 5, Gate/Sync/Clock Description, for additional information about the
use and programming of gate and sync signals.
Rev.: B.4
Page 60
Page 60Pentek Model 6230/6231 Operating Manual
3.14Interrupt Mask Register
The Interrupt Mask Register contains one enable bit for each interrupt condition
defined for the controlling processor. Each processor on the VIM baseboard has its
own version of this register.
The following table shows which bit in this register is associated with each interrupt
condition. Table 3−18, on the next page, provides description of the interrupt condition
associated with each of these bits.
All bits default to the logic '0' state at power on and reset
Each bit of this register enables or disables the generation of the interrupt to that pro−
cessor. Setting the bit associated with a given interrupt to the logic '1' state enables the
generation of an interrupt to the processor when that interrupt condition occurs. When
a bit is cleared to logic '0', the processor will not be interrupted by the associated inter−
rupt condition.
The following two sections, Interrupt Flag Register, Section 3.15, and Status Register,
Section 3.16, contain associated functions for using these processor interrupts.
Rev.: B.4
Page 61
Pentek Model 6230/6231 Operating ManualPage 61
3.14Interrupt Mask Register (continued)
Table 3−18: Interrupt Register Bits
Bit NameBit PositionInterrupt Function
Each of these bits is associated with a Temperature/Voltage
sensor interrupt for part of the Model 6230/6231. (There are four
TEMP 2
TEMP 1
VALID CLKD10
CLK LOSSD09This bit is associated with a Clock loss interrupt.
GATE D
GATE C
GATE B
GATE A
SYNCD04
OVLD CH4
OVLD CH3
OVLD CH2
OVLD CH1
D12
D11
D08
D07
D06
D05
D03
D02
D01
D00
sensors on the Model 6230 [two on the Model 6231]. These are
controlled using the Hardware Monitor Port. To determine the
sensor causing the interrupt, use the Hardware Monitor Port
Register, Section 3.6.)
This bit is associated with a Valid Clock indication interrupt. The
interrupt status will go to logic '1' 16 clock cycles after detection
of a clock. It will also indicate that 16 clock cycles have occurred
since the beginning of a BIFO reset. This can be used as an
indication that the BIFO reset is complete, since a BIFO reset
requires several clock cycles to complete. The bit will remain
high during normal operation, thus it should be masked
(disabled) after it is no longer useful to prevent undesired
interrupts.
Each of these bits is associated with a Gate interrupt. The
interrupt occurs at the start or end of the respective gate signal,
depending on the setting of INT EDGE in the Gate Control
register (Section 3.10.1).
This bit is associated with a Sync interrupt. The interrupt occurs
at the start of the sync pulse.
Each of these bits is associated with an interrupt for an analog
input Overload indication from one of the A/D converters (A/D4,
A/D3, A/D2, or A/D1, respectively).
Note: These bit definitions apply to the Interrupt Mask Register, Section 3.14, Interrupt Flag
Register, Section 3.15, and Interrupt Status Register, Section 3.16
Rev.: B.4
Page 62
Page 62Pentek Model 6230/6231 Operating Manual
3.15Interrupt Flag Register
The Interrupt Flag Register has one read/clear bit associated with each interrupt con−
dition for the VIM processor (the same bit associations as the Interrupt Mask Register,
Section 3.14). Each processor on the VIM baseboard has its own version of this register.
The following table shows which bit in this register is associated with each interrupt
condition. Table 3−18, on the prior page, provides description of the interrupt condit ion
associated with each of these bits.
Table 3−19: Interrupt Flag Register
R/Clr @ VIM_base+0x0048
D15 – D13D12D11D10D09D08
Bit Name
Function
Bit Name
Function
IMPORTANT! When reset, including power−up, the state of this register is unknown.
Not usedTEMP 2TEMP 1VALID CLKCLK LOSSGATE D
Write with zeros,
Mask when reading
Read:
Clear:
0 = No interrupt
1 = Interrupt latched
0 = Enabled to latch
1 = Clear latch
You should clear this register before using it, by writing '1's into all bits.
0 = No interrupt
1 = Interrupt latched
0 = Enabled to latch
1 = Clear latch
Each bit of this register latches an interrupt occurrence. A logic '1' in any bit in this
register indicates that an interrupt has occurred. Note that when any bit in the Status
Register (Section 3.16) changes to logic '1', the corresponding bit in this register will also
be set to logic '1'. However, when a bit in the Status Register changes from logic '1' to
logic '0', the corresponding latched bit in this register does not clear, but remains at the
logic '1' state.
Since these bits latch in response to an interrupt, to detect subsequent interrupts, you
must clear the bits in this register. To clear any bit in this register that is set to the logic
'1' state, you must write to that bit twice, first with a '1' to clear the bit, then with a '0' to
re−enable the bit for latching.
Rev.: B.4
Page 63
Pentek Model 6230/6231 Operating ManualPage 63
3.16Interrupt Status Register
The Interrupt Status Register has one read−only bit associated with each interrupt con−
dition for the VIM processor (the same bit associations as the Interrupt Mask Register,
Section 3.14). Each processor on the VIM baseboard has its own version of this register.
The following table shows which bit in this register is associated with each interrupt
condition. Table 3−18, on page 61, provides description of the interrupt condition asso−
ciated with each of these bits.
When reset or power−up, the state of this register is unknown.
0 = No
interrupt
1 = Gate
0 = No
interrupt
1 = Gate
0 = No
interrupt
1 = Sync
0 = No
interrupt
1 = Overload
0 = No
interrupt
1 = Overload
0 = No
interrupt
1 = Overload
0 = No
interrupt
1 = Overload
Each bit in this register changes whenever the associated interrupt indication changes.
A bit changes to the logic '1' state whenever that interrupt occurs, and clears to logic '0'
state when that interrupt clears (whereas the associated bit in the Interrupt Flag Regis−
ter, Section 3.15, remains latched at logic '1' until it is cleared by that register’s clear
action).
If the corresponding bit in the Interrupt Mask Register (Section 3.14) has been set to the
logic '1' state, then a Status Register bit transition from logic '0' to logic '1' will also gen−
erate an interrupt to the processor on the VIM baseboard. If you do not want an inter−
rupt to occur and have set the corresponding interrupt mask bit to '0', you may poll this
Status Register bit to determine if such an event has occurred.
Rev.: B.4
Page 64
Page 64Pentek Model 6230/6231 Operating Manual
3.17Semaphore Register
The Semaphore Register contains communications bits that may be written by one pro−
cessor on the VIM baseboard and read by all four processors [two processors for the
Model 6231]. This register is common to all processors on the baseboard.
The following table shows which bits are associated with each processor for writing.
Table 3−21: Semaphore Register
R/W @ VIM_base+0x0050
D15D14D13D12D11D10D09D08
Bit Name
Function
D3D2D1D0C3C2C1C0
Bits written by Processor DBits written by Processor C
D07D06D05D04D03D02D01D00
Bit Name
Function
* For a Model 6231 installed in the bottom VIM position, D07−D04 = Processor D, D03−D00 = Processor C.
B3B2B1B0A3A2A1A0
Bits written by Processor B*Bits written by Processor A*
All bits default to the logic '0' state at power on and reset
Each processor may write only four of the bits in this register, but any processor may
read all bits. These bits allow coordination between programs running in the base−
board processors.
NOTE:
Bits D15 through D08 are not used on the Model 6231.
Rev.: B.4
Page 65
Pentek Model 6230/6231 Operating ManualPage 65
3.18I/O Direction 2 Register (Model 6231 only)
The I/O Direction 2 Register determines the input/output direction of each bit in the
FPGA I/O Data 2 Register, Section 3.19 below. Each data bit can be individually set for
input or output. This register is accessible only to Processor A on the VIM baseboard
(or Processor C for a Model 6231 that is installed in the bottom VIM position on the
baseboard).
Bit Name
Function
NOTE:
This register is available only on the Model 6231 with Option 102.
The following table shows which bit of this register is associated with each FPGA I/O
data bit, and the paragraph following the table provides descriptions of these bits.
All bits default to the logic '0' state at power on and reset
0 = Input bit
1 = Output bit
Refer to the connector diagram in Table 2−6, Section 2.4.5, for the pin assignments of
each bit.
When a bit is cleared to logic '0' (its default state), the corresponding data bit can be
input at the front panel FPGA connector (Section 2.4.5). When a bit is set to logic '1', the
corresponding data bit is set for output.
Rev.: B.4
Page 66
Page 66Pentek Model 6230/6231 Operating Manual
3.19I/O Data 2 Register (Model 6231 only)
The I/O Data 2 Register allows you to read or write data from the Virtex FPGA to the
front panel FPGA connector (Section 2.4.5). The direction (read or write) for each bit is
controlled by the corresponding bit of the I/O Direction Register, Section 3.18 above.
This register is accessible only to Processor A on the VIM baseboard (or Processor C for
a Model 6231 that is installed in the bottom VIM position on the baseboard).
Bit Name
Function
NOTE:
This register is available only on the Model 6231 with Option 102.
The following table shows which bit in this register is associated with each connector
data bit.
When reset or power−up, the state of this register is unknown.
FPGA
I/O 23
FPGA
I/O 22
Each bit can be used as data Input or Output, depending on
corresponding bit of I/O Direction 2 Register, Section 3.18
FPGA
I/O 21
FPGA
I/O 20
FPGA
I/O 19
FPGA
I/O 18
FPGA
I/O 17
FPGA
I/O 16
These data bits are always enabled. Refer to the connector pin diagram in Table 2−6,
Section 2.4.5, for the pin assignments of each bit.
Rev.: B.4
Page 67
Pentek Model 6230/6231 Operating ManualPage 67
3.20I/O Direction Register
The I/O Direction Register determines the input/output direction of each bit in the
FPGA I/O Data Register (Section 3.22). Each data bit can be individually programmed
for input or output. Processor A and Processor C on the VIM baseboard have their own
version of this register, for the FPGA associated with that processor.
This register is available on the Model 6230 and only with Option 102 on the Model
6231.
The following table shows which bit of this register is associated with each FPGA I/O
data bit.
Table 3−24: I/O Direction Register
R/W @ VIM_base+0x0060 (Proc A & C only)
D15D14D13D12D11D10D09D08
Bit Name
Function
Bit Name
Function
DIR15DIR14DIR13DIR12DIR11DIR10DIR9DIR8
0 = Input bit
1 = Output bit
D07D06D05D04D03D02D01D00
DIR7DIR6DIR5DIR4DIR3DIR2DIR1DIR0
0 = Input bit
1 = Output bit
All bits default to the logic '0' state at power on and reset
Refer to the connector diagram in Table 2−6, Section 2.4.5, for the pin assignments of
each bit.
When a bit is cleared to logic '0' (its default state), the corresponding data bit can be
input at the front panel FPGA connector (Section 2.4.5). When a bit is set to logic '1', the
corresponding data bit is set for output.
Rev.: B.4
Page 68
Page 68Pentek Model 6230/6231 Operating Manual
3.21I/O Enable Register
The I/O Enable Register enables connection of the I/O bits of the I/O Data Register
(Section 3.22) to the front panel FPGA connector (Section 2.4.5). Processor A and Pro−
cessor C on the VIM baseboard have their own version of this register, for the FPGA
associated with that processor (for example, FPGA1 for Processor A).
This register is available on the Model 6230 and only with Option 102 on the Model
6231.
The following table shows the contents of the I/O Enable Register. There are two
enable bits that allow you to enable either or both 8−bit halves of the 16−bit data word.
The sections following the table give descriptions of the bits in this register.
Table 3−25: I/O Enable Register
R/W @ VIM_base+0x0064 (Proc A & C only)
D15 – D02D01D00
Bit Name
Function
Not usedI/O EN1I/O EN2
Write with zeros,
Mask when reading
All bits default to the logic '0' state at power on and reset
0 = Disable
1 = Enable
3.21.1I/O EN1Bit D01
On the Model 6230, this bit enables connection of I/O bits D7 through D0 of
the Data Register to the front panel FPGA connector. On the Model 6231,
this bit enables connection of I/O bits D15 through D0. When this bit is set
to logic '1', these I/O bits of the Data Register are enabled at the front panel.
When the bit is cleared to logic '0', they are disabled.
3.21.2I/O EN2Bit D00
On the Model 6230, this bit enables connection of I/O bits D15 through D8
of the Data Register to the front panel FPGA connector. On the Model 6231,
this bit enables connection of I/O bits D23 through D16. When this bit is set
to logic '1', these I/O bits of the Data Register are enabled at the front panel.
When the bit is cleared to logic '0', they are disabled.
Rev.: B.4
Page 69
Pentek Model 6230/6231 Operating ManualPage 69
3.22I/O Data Register
The I/O Data Register allows you to read or write data from the Virtex FPGA to the
front panel FPGA connector (Section 2.4.5). The direction (read or write) for each bit is
controlled by the corresponding bit of the I/O Direction Register, Section 3.20. The
data bits are enabled or disabled for connection to the front panel connector by the I/O
Enable Register, Section 3.21. Processor A and Processor C on the VIM baseboard have
their own version of this register, for the FPGA associated with that processor (for
example, FPGA1 for Processor A).
This register is available on the Model 6230 and only with Option 102 on the Model
6231.
The following table shows which bit in this register is associated with each connector
data bit.
Table 3−26: I/O Data Register
R/W @ VIM_base+0x0068 (Proc A & C only)
D15D14D13D12D11D10D09D08
Bit Name
Function
Bit Name
Function
FPGAn
I/O 15
D07D06D05D04D03D02D01D00
FPGAn
I/O 7
In Bit Name, ‘n’ is the FPGA number (1 or 2) associated with the processor.
When reset or power−up, the state of this register is unknown.
FPGAn
I/O 14
Each bit can be used as data Input or Output, depending on
corresponding bit of I/O Direction Register, Section 3.20
FPGAn
I/O 6
Each bit can be used as data Input or Output, depending on
corresponding bit of I/O Direction Register, Section 3.20
FPGAn
I/O 13
FPGAn
I/O 5
FPGAn
I/O 12
FPGAn
I/O 4
FPGAn
I/O 11
FPGAn
I/O 3
FPGAn
I/O 10
FPGAn
I/O 2
FPGAn
I/O 9
FPGAn
I/O 1
FPGAn
I/O 8
FPGAn
I/O 0
In the Bit Name of each bit in this table, ‘n’ is the FPGA number (for example, for Pro−
cessor A, bit D15 is ‘FPGA1 I/O 15’). [For the Model 6231, there is no FPGA number.]
Refer to the connector pin diagram in Table 2−6, Section 2.4.5, for the pin assignments of
each bit.
Rev.: B.4
Page 70
Page 70Pentek Model 6230/6231 Operating Manual
3.23Graychip 0 & 1 Registers
The Graychip Registers allow you to read or write data from/to each Graychip GC4016
DDR. There are eight sets of registers on the Model 6230 [four sets on the Model 6231],
one for each GC4016 on the module, with 128 registers in each set. Each processor on
the VIM baseboard can use two sets of registers, for the two GC4016 DDRs associated
with that processor. The register addresses determines which CG4016 DDR to read/
write to — use register set ‘Graychip 0’ for the first CG4016, use register set ‘Graychip
1’ for the second CG4016. Table 3−28, at the bottom of this page, identifies the CG4016s
DDRs that can be set by each processor and the register set to use for each CG4016. The
following table shows the bit layout of each register of these sets. You can read or
write one byte at a time to each register.
Table 3−27: Graychip 0 & 1 Registers
Graychip 0 Register Set − R/W @ VIM_base+0x0100 – 017C
D15 – D08D07D06D05D04D03D02D01D00
Bit Name
Function
Bit Name
Function
Not usedD7D6D5D4D3D2D1D0
Write zeros,
Mask read
Graychip 1 Register Set − R/W @ VIM_base+0x0180 – 01FC
Eight bits of data to/from GC4016
D15 – D08D07D06D05D04D03D02D01D00
Not usedD7D6D5D4D3D2D1D0
Write zeros,
Mask read
When reset or power−up, the state of these registers is unknown.
Eight bits of data to/from GC4016
Table 3−28: Processor/Graychip Register Sets
ProcessorGraychipRegister Set
Processor A *
Processor B *
Processor C
Processor D
* Processor C and Processor D, respectively, for a Model 6231
Graychip GC4016 DDR library functions are provided in the ReadyFlow board support
software package for the Model 6230 (Pentek part #801.62300). Refer to the Graychip
GC4016 data sheet, Appendix C, for detailed information on these registers.
Rev.: B.4
Page 71
Pentek Model 6230/6231 Operating ManualPage 71
Chapter 4: Data Formatting and Routing
4.1Overview
This chapter provides descriptions of the data routing and formatting for the Model
6230/6231.
4.2Data Routing and Formats
Each analog input channel includes an AD6645 A/D converter that can provide data to
any of the GC4016 DDRs. Each FPGA is configured to accept 16 channels of serial data
from four GC4016s, or A/D data directly from two AD6645s in ‘DDR Bypass’ modes.
Use the Channel Control Register (Section 3.12) to select one of the seven Data Routing
modes and format of the data from the FPGA to the VIM baseboard processor BIFOs.
In ‘DDR Bypass’ modes (Sections 4.2.1 to 4.2.3), each FPGA takes the A/D data directly,
bypassing the GC4016 DDRs. The output is written to the VIM BIFOs at a programmed
decimation rate, and may be unpacked, packed by time of arrival, or packed by channel.
In ‘DDR’ routing modes (Sections 4.2.4 to 4.2.7), each VIM BIFO can receive data from
only the eight receiver channels of the two GC4016s associated with that processor (DSP
CH1 = Channel A output of first GC4016 ... DSP CH8 = Channel D output of second
GC4016; see also Table 3−12 on page 52). This can be expanded to 16 or more channels
with additional user programming. The channels are scanned by the FPGA’s data for−
matter state machine starting at the first channel. You may enable or disable any number
of channels using the Channel Enable Register (Section 3.9). If you disable a channel, its
data is skipped in the scan. When the gate is disabled, all BIFO writes are disabled. The
data formatter state machine is reset when the gate is disabled, but will complete scan−
ning through the eight receiver channels and writing them to the BIFO before stopping.
All channels enabled for output to the BIFO MUST
guarantee that the eight channels are from the same data sample, the GC4016 pair must
be synchronized. Output with identical rates but not synchronized is possible, but the
relationship between data from different channels is not known.
For ‘DDR’ modes, ensure the Graychip GC4016s are configured accordingly using the
GC4016 registers (Section 3.23). The serial−to−parallel converters in the FPGAs accept the
GC4016 serial data at the full clock rate, and the SCLK output from the GC4016s is not
used, so the GC4016s must be set up for SERIAL output at the full clock rate. The FPGA
converters expect the data from the GC4016s to be configured as one SFS pulse per I/Q
pair, with the data word size either 16 or 24 bits per I or Q. You must set the mode bits in
the Channel Control Register (Section 3.12) to the data word size defined for the
GC4016s.
be at the same rate. Additionally, to
The ‘DDR’ modes assume that the GC4016s are set up for COMPLEX output,
providing both I and Q values. If the GC4016s are configured for REAL output, data
Note
is still output from the GC4016s as two samples per channel. The value in each
sample is real only and both samples are identical, however, the FPGA still marks
the data as I and Q samples.
Rev.: B.4
Page 72
Page 72Pentek Model 6230/6231 Operating Manual
4.2Data Routing and Formats (continued)
4.2.1DDR Bypass Mode, A/D Data, Unpacked (001)
This mode takes the raw 14−bit data directly from the two A/Ds, bypassing
the GC4016 DDRs. The output is written to the VIM BIFO at a programmed
decimation rate N, where N is 1 to 4096, only when the gate is enabled.
Each processor BIFO receives data from only one input channel: A/D 1 for
Processor A, A/D 2 for Processor B, A/D 3 for Processor C, and A/D 4 for
Processor D. Each sample is left justified in the 14 most significant bits in
both of the 16−bit halves of the 32−bit BIFO data word. Identical data is
placed in both halves of the word. Data is in the following format:
Table 4−1: Output Data Format − DDR Bypass Mode, Unpacked
d = raw A/D data bit (same value in both halves of word)
4.2.2DDR Bypass Mode, A/D Data, Time Packed (010)
This mode takes the raw 14−bit data directly from the two A/Ds, bypassing
the GC4016 DDRs. The output is written to the VIM BIFO at a programmed
decimation rate N/2, where N is 1 to 4096, only when the gate is enabled.
Each processor BIFO receives data from only one input channel: A/D 1 for
Processor A, A/D 2 for Processor B, A/D 3 for Processor C, and A/D 4 for
Processor D. Each sample is left justified in the 14 most significant bits in
each of the 16−bit halves of the 32−bit BIFO data word. Data is packed with
input data word data(t) in the upper half of the 32 bits, and word data(t−1)
in the lower half of the 32 bits. Data is in the following format:
Table 4−2: Output Data Format − DDR Bypass Mode, Time Packed
This mode takes the raw 14−bit data directly from the two A/Ds, bypassing
the GC4016 DDRs. The output is written to the VIM BIFO at a programmed
decimation rate N, where N is 1 to 4096, only when the gate is enabled.
Data is packed with data from A/D 1 in the upper half of the 32−bit word,
and data from A/D 2 in the lower half of the 32−bit word for processor A
and B BIFOs, and with data from A/D 3 in the upper half of the 32−bit word,
and data from A/D 4 in the lower half of the 32−bit word for processor C
and D BIFOs. Each sample is left justified in the 14 most significant bits in
each of the 16−bit halves of the 32−bit data word.
Data is in the following format:
Table 4−3: Output Data Format − DDR Bypass Mode, Channel Packed
You MUST set the decimation rate dividers of both processor channels (A
and B, or C and D) at the same rate and synchronized for this mode to
function properly!
= raw data from A/D 2 or A/D 4
2−4
00
Rev.: B.4
Page 74
Page 74Pentek Model 6230/6231 Operating Manual
4.2Data Routing and Formats (continued)
4.2.4DDR Mode, 16−Bit, Unpacked I/Q, Tagged (100)
This mode expects data from the two GC4016 DDRs to be 32−bit I/Q pairs.
Data is accepted and multiplexed to the VIM BIFO in a scanned manner
starting with the first of the eight receiver channels. The formatter state
machine is reset when the gate is disabled and always starts at the first
channel when the gate is enabled. Any disabled channels are skipped. BIFO
writes are disabled except when the gate is enabled.
The data is output in two consecutive 32−bit words per I/Q pair—the first
word is the I data word. The FPGA inserts a DDR channel number tag in the
lowest four bits of each 32−bit word. Data is in the following format:
Table 4−4: Output Data Format − DDR Mode, 16−bit, Unpacked I/Q, Tagged
X = Channel # Tag (DSP CH1 = 0000 ... DSP CH8 = 0111)
NOTE:
The DDR modes assume that the GC4016s are set up for
COMPLEX output (the default output), providing both I and Q
values. If the GC4016s are configured for REAL output, data is still
output from the GC4016s as two samples per channel. The value
in each sample is real only and both samples are identical,
however, the FPGA still marks the data as I and Q samples.
Rev.: B.4
Page 75
Pentek Model 6230/6231 Operating ManualPage 75
4.2Data Routing and Formats (continued)
4.2.5DDR Mode, 24−Bit, Unpacked I/Q, Tagged (101)
This mode expects data from the two GC4016 DDRs to be 48−bit I/Q pairs.
Data is accepted and multiplexed to the VIM BIFO in a scanned manner
starting with the first of the eight receiver channels. The formatter state
machine is reset when the gate is disabled and always starts at the first
channel when the gate is enabled. Any disabled channels are skipped. BIFO
writes are disabled except when the gate is enabled.
The data is output in two consecutive 32−bit words per I/Q pair—the first
word is the I data word. The FPGA inserts a DDR channel number tag in the
lowest four bits of each 32−bit word. Data is in the following format:
Table 4−5: Output Data Format − DDR Mode, 24−bit, Unpacked I/Q, Tagged
X = Channel # Tag (DSP CH1 = 0000 ... DSP CH8 = 0111)
NOTE:
The DDR modes assume that the GC4016s are set up for
COMPLEX output (the default output), providing both I and Q
values. If the GC4016s are configured for REAL output, data is still
output from the GC4016s as two samples per channel. The value
in each sample is real only and both samples are identical,
however, the FPGA still marks the data as I and Q samples.
Rev.: B.4
Page 76
Page 76Pentek Model 6230/6231 Operating Manual
4.2Data Routing and Formats (continued)
4.2.6DDR Mode, 16−Bit, Packed I/Q (110)
This mode expects data from the two GC4016 DDRs to be 32−bit I/Q pairs.
Data is accepted and multiplexed to the VIM BIFO in a scanned manner
starting with the first of the eight receiver channels. The formatter state
machine is reset when the gate is disabled and starts always at the first
channel when the gate is enabled. Any disabled channels are skipped. BIFO
writes are disabled except when the gate is enabled.
The output data format is a 16−bit I and a 16−bit Q data value packed into a
32−bit word. Since there are no unused bits, data is not tagged for channel
identification, but guaranteed to begin with the first enabled channel once
gate is enabled. The user must then know which channels are skipped
because they are disabled to determine what channel data belongs to.
Data is in the following format:
Table 4−6: Output Data Format − DDR Mode, 16−bit, Packed I/Q
GC4016−formatted tagging can be done in this mode by configuring the
GC4016 registers—however, if GC4016 tagging is used, the least significant
four bits of each value, D3 to D0 and D16 to D19, will not be available as
data bits.
NOTE:
The DDR modes assume that the GC4016s are set up for
COMPLEX output (the default output), providing both I and Q
values. If the GC4016s are configured for REAL output, data is still
output from the GC4016s as two samples per channel. The value
in each sample is real only and both samples are identical,
however, the FPGA still marks the data as I and Q samples.
Rev.: B.4
Page 77
Pentek Model 6230/6231 Operating ManualPage 77
4.2Data Routing and Formats (continued)
4.2.7DDR Mode, 24−Bit, Packed I/Q (111)
This mode expects data from the two GC4016 DDRs to be 48−bit I/Q pairs.
Data is multiplexed to the VIM BIFO in a scanned manner starting with the
first of the eight receiver channels. The formatter state machine is reset
when the gate is disabled and always starts at the first channel when the
gate is enabled. Any disabled channels are skipped. BIFO writes are dis−
abled except when the gate is enabled. Also, in this mode, the lowest deci−
mation rates cannot be reached because the minimum clocks between
samples are 48 instead of 32.
The output data format is a 16−bit I and a 16−bit Q data value packed into a
32−bit word, where the 16 bits are the most significant 16 of the 24−bit input
words. Since there are no unused bits, data is not tagged for channel identi−
fication, but guaranteed to begin with the first enabled channel once the gate
is enabled. The user must then know which channels are skipped because
they are disabled to determine what channel data belongs to.
Data is in the following format:
Table 4−7: Output Data Format − DDR Mode, 16−bit, Packed I/Q
The DDR modes assume that the GC4016s are set up for
COMPLEX output (the default output), providing both I and Q
values. If the GC4016s are configured for REAL output, data is still
output from the GC4016s as two samples per channel. The value
in each sample is real only and both samples are identical,
however, the FPGA still marks the data as I and Q samples.
Rev.: B.4
Page 78
Page 78Pentek Model 6230/6231 Operating Manual
This page is intentionally blank
Rev.: B.4
Page 79
Pentek Model 6230/6231 Operating ManualPage 79
Chapter 5: Timing and Synchronization
5.1Overview
This chapter provides descriptions of the timing and synchronization of the module
processing. The Model 6230/6231 provides four gates, one sync, and one clock. These
signals go out on the LVDS sync bus if the board is a bus Master, or come in on the
LVDS sync bus for a bus Slave unit. The Model 6230/6231 may, alternately, bypass the
sync bus and use its own gate, sync, or clock sources for timing and control.
See Figures 5−1 and 5−2, below and on the next page, for logic diagrams of these timing
signals.
GATES
A−D
Front Panel
SYNC Bus
MCLK
SYNC
4
4
GATE SRC
LVDS
Receiver
LVDS
Driver
MASTR
Master Contr ol, D0
Gate Control, D4
CLK SRC
Master Control, D4
SYNC SRC
Master Control, D9
Gate/Trigger
44
Master Clock
Virtex−E
FPGA (2)
Circuit
Divider
CLK IN
CLK IN
SYNC
AD6644
A/Ds (4)
GC4016
DDRs (8)
EXT SYNC EN
Master Contr ol, D7
Gate Control , D0
Master Contr ol, D2
EXT GATE
EXT CLK
ENABLE
CLK IN A
FIFO
CLK IN B
Quad Processor Board
GATE A A)(Processor
GATE B B)(Processor
GATE C C)(Processor
GATE D D)(Processor
Figure 5−1: Model 6230 Gate/Sync/Clock Logic
Gate Input Select
ENABLE
CLK IN C
FIFO
Sync/Gate Gen, D0
SYNC A (Proc A)
SYNC B B)(Proc
SYNC C C)(Proc
SYNC D D)(Proc
Sync/Gate Gen, D1
64 MHz
Oscillator
ENABLE
FIFO
ENABLE
CLK IN D
FIFO
EXT TTL
SYNC IN
EXT TTL
GATE/TRIG IN
Front Panel
Connectors
EXT CLK IN
Rev.: B.4
Page 80
Page 80Pentek Model 6230/6231 Operating Manual
5.1Overview (continued)
GATES C,D
GATES A,B
Front Panel
SYNC Bus
MCLK
SYNC
2
2
GATE SRC
Gate Control, D4
LVDS
Receiver
2
2
LVDS
Driver
CLK SRC
Master Control, D4
SYNC SRC
Master Control, D9
2
Gate/Trigger
2
2
Virtex−E
FPGA
Circuit
(C & D)
(A & B)
CLK IN
CLK IN
SYNC
AD6644
A/Ds (2)
GC4016
DDRs (4)
EXT SYNC EN
Master Contr ol, D7
EXT GATE
Gate Control , D0
Gate Input Select
ENABLEENABLE
CLK IN
(A or C)
FIFO
CLK IN
(B or D)
FIFO
Quad Processor Board
Sync/ Gate Gen , D0
SYNC A (Proc A or C)
SYNC B (Proc B or D)
Sync/ Gate Gen , D1
GATE A (Proc A or C)
GATE B (Proc B or D)
EXT TTL
SYNC IN
EXT TTL
GATE/TRG IN
Front Panel
MASTR
Master Contr ol, D0
Master Clock
Divider
EXT CLK
Master Contr ol, D2
64 MHz
Oscillator
Connectors
EXT CLK IN
5.2Sync
The sync signal is used for data synchronization by the CG4016 DDRs. The sync can be
driven either from the front panel LVDS sync bus or from an external TTL sync input
(Sections 2.4.3 and 2.4.4), or a sync signal can be generated by a register write by each
processor (Section 3.13). When the Model 6230/6231 is a sync bus Master, the gener−
ated sync is output to the LVDS Sync Bus.
5.3Clock
The clock for all the board functions can be driven from either the front panel LVDS
sync bus or an external clock input (Section 2.4.2), or a clock can be generated using the
on−board 64−MHz oscillator (Section 3.7.7). When the Model 6230/6231 is a sync bus
Master, the on−board clock is output to the LVDS Sync Bus.
Figure 5−2: Model 6231 Gate/Sync/Clock Logic
Rev.: B.4
Page 81
Pentek Model 6230/6231 Operating ManualPage 81
5.4Gates
The four gates are used to enable writes to the VIM baseboard BIFOs. Each gate can be
driven either from the front panel LVDS sync bus or from an external TTL gate input
(Sections 2.4.3 and 2.4.4), or each gate may be generated from a register write by each
processor (Section 3.13). Any receiver channel may use any one of the four gate sources
as its gate, but each gate is generated by an individual processor channel on a sync bus
Master board.
Gates may be disabled on a channel−by−channel basis, in which case BIFO writes are
always enabled. The polarity of the external gate source is programmable. All gate
features are programmable on a channel−by−channel basis, except external gate polar−
ity, which is selectable from Processor A only and is global (Section 3.10). Gate transi−
tions from high to low or low to high can programmably create interrupts on any
processor.
When the Model 6230/6231 is a sync bus Master, the generated gates are output to the
LVDS Sync Bus. When the Model 6231 is a Master, the board generates four gates for
the LVDS sync bus: Gates A and C from Processor A (or Processor C for a Model 6231
that is installed in the bottom VIM position on the baseboard), and Gates B and D from
Processor B (or Processor D for a 6231 in the bottom VIM position).
The external TTL gate or the register write gate may be programmed to act as a trigger
(Trigger mode). In this case, the gate is generated after a desired polarity logic transi−
tion on the external gate source or register write, and the resulting gate continues either
indefinitely (Hold mode) or for a programmed number of BIFO writes up to 16,383
(Trigger Length). At any time the triggered gate may be asynchronously disabled by a
control register write.
Rev.: B.4
Page 82
Page 82Pentek Model 6230/6231 Operating Manual
This page is intentionally blank
Rev.: B.4
Page 83
Pentek Model 6230/6231 Operating ManualPage A−1
Appendix A: Configuration EEPROM Format
A.1Introduction
All VIM Modules contain an Identification (ID) EEPROM with a checksum stored in the
last location. When booted, each VIM baseboard processor checks for the presence of a
VIM module, through the presence bit on the VIM connector. If a module is present,
the processor reads the module’s ID EEPROM and generates a 16−bit checksum from
the data read. If the generated checksum does not match the stored value, the proces−
sor illuminates its red LED. Even though the processor illuminates LED 0, the red LED,
the processor is fully functional.
The contents of the ID EEPROM, plus the calculated checksum, are stored in Global
SDRAM on the VIM baseboard. Refer to the respective VIM baseboard Operating Man−
ual for the location of the Global SDRAM. The first word stored is a valid data flag,
based on the checksum: valid checksum (
no VIM module installed (
0xFFFF FFFE).
0x00EE C0DE), bad checksum (0x000B ADC5), or
The ID EEPROM is accessed using a Model 6230/6231 register at memory address
VIM_base+0x0000, as offset from the VIM base address defined for the applicable VIM
baseboard. The bit layout of this register is defined in Table A−1, below.
Table A−1: VIM ID EEPROM Register
R/W @ VIM_base+0x0000
Bit #D31 − D4D3D2D1D0
Bit NameNot UsedChip SelectSerial Data OutReservedSerial Clock
Table A−2 on the next page shows the contents of the ID EEPROM for the Model 6230,
and an explanation of what each pair of 16−bit words is used for. The SDRAM location
is the address where each pair of words is stored relative to the start address of the
Global SDRAM on the VIM baseboard.
Rev. B.4
Page 84
Page A−2Pentek Model 6230/6231 Operating Manual
A.2EEPROM Format Example (continued)
Table A−2: EEPROM Example (Model 6230 shown)
EEPROM
Word Location
00/01+0x000x00EE C0DEValid data flag
02/03+0x040x6230 0000Model, Model Extension (Extension currently unused)
04/05+0x080x0354 2001PCB number − this is taken from the VIM board
Rev A = 0001, Rev B = 0002, etc.;
Option #1 − up to 9 installed options may be
specified, if there are no options, field = 'FFFF'
Module−specific code goes in locations 16 through 61
Word 62 should always be cleared (0000).
Word 63 contains the checksum of words 0 − 61.
Rev. B.4
Page 85
Pentek Model 6230/6231 Operating ManualPage B−1
Appendix B: Analog Devices AD6644 A/D Converter
B.1Introduction
The following pages are a reprint of the Analog Devices AD6644 14−Bit, 65 MHz A/D
Converter data sheet.
Rev.: B.4
Page 86
Page B−2Pentek Model 6230/6231 Operating Manual
This page is intentionally blank
Rev.: B.4
Page 87
14-Bit, 40 MSPS/65 MSPS
a
FEATURES
65 MSPS Guaranteed Sample Rate
40 MSPS Version Available
Sampling Jitter < 300 fs
100 dB Multitone SFDR
1.3 W Power Dissipation
Differential Analog Inputs
Digital Outputs
Two’s Complement Format
3.3 V CMOS-Compatible
Data Ready for Output Latching
APPLICATIONS
Multichannel, Multimode Receivers
AMPS, IS-136, CDMA, GSM, Third Generation
Single Channel Digital Receivers
Antenna Array Processing
Communications Instrumentation
Radar, Infrared Imaging
Instrumentation
PRODUCT DESCRIPTION
The AD6644 is a high-speed, high-performance, monolithic
14-bit analog-to-digital converter. All necessary functions,
including track-and-hold (T/H) and reference, are included onchip to provide a complete conversion solution. The AD6644
provides CMOS-compatible digital outputs. It is the third generation in a wideband ADC family, preceded by the AD9042 (12-bit
41 MSPS) and the AD6640 (12-bit 65 MSPS, IF sampling.)
A/D Converter
AD6644
Designed for multichannel, multimode receivers, the AD6644 is
part of ADI’s new SoftCell™ transceiver chipset. The AD6644
achieves 100 dB multitone, spurious-free dynamic range (SFDR)
through the Nyquist band. This breakthrough performance eases
the burden placed on multimode digital receivers (software radios)
which are typically limited by the ADC. Noise performance is
exceptional; typical signal-to-noise ratio is 74 dB.
The AD6644 is also useful in single channel digital receivers
designed for use in wide-channel bandwidth systems (CDMA,
W-CDMA). With oversampling, harmonics can be placed outside the analysis bandwidth. Oversampling also facilitates the use of
decimation receivers (such as the AD6620), allowing the noise
floor in the analysis bandwidth to be reduced. By replacing traditional analog filters with predictable digital components, modern
receivers can be built using fewer “RF” components, resulting
in decreased manufacturing costs, higher manufacturing yields,
and improved reliability.
The AD6644 is built on Analog Devices’ high-speed complementary bipolar process (XFCB) and uses an innovative, multipass
circuit architecture. Units are packaged in a 52-terminal LowProfile Quad Plastic Flatpack (LQFP) specified from –25°C
to +85°C.
PRODUCT HIGHLIGHTS
1. Guaranteed sample rate is 65 MSPS.
2. Fully differential analog input stage.
3. Digital outputs may be run on 3.3 V supply for easy interface
to digital ASICs.
4. Complete Solution: reference and track-and-hold.
5. Packaged in small, surface-mount, plastic, 52-terminal LQFP.
AVCCDV
AIN
AIN
V
2.4V
REF
ENCODE
ENCODE
SoftCell is a trademark of Analog Devices, Inc.
INTERNAL
TIMING
CC
GNDD8D9D10D11D12D13DRYOVRDMIDD0D1D2D3D4D5D6D7
REV. 0
Information furnished by Analog Devices is believed to be accurate and
reliable. However, no responsibility is assumed by Analog Devices for its
use, nor for any infringements of patents or other rights of third parties
which may result from its use. No license is granted by implication or
otherwise under any patent or patent rights of Analog Devices.
ENCODE to DATA Delay (Hold Time) is the absolute minimum propagation delay through the analog-to-digital converter.
4
ENCODE to DATA Delay (Setup Time) is calculated relative to 65 MSPS (50% duty cycle). In order to calculate t
Newt
5
DRY is an inverted and delayed version of the encode clock. Any change in the duty cycle of the clock will correspondingly change the duty cycle of DRY.
6
Data Ready to DATA Delay(t
and t
Newt
Newt
Specifications subject to change without notice.
H_DR
= (t
S_DR
S_DR
= t
S_E
ENC(NEW)
for a given encode use the following equations:
S_DR
= t
H_DR
ENC(NEW)
= t
S_DR
ENC(NEW)
)/DATA, OVR
– % Change(t
– % Change(t
– t
+ t
ENC
/2 – t
ENCH
/2 – t
ENCH
2
and t
ENC
)) × t
ENCH
)) × t
ENCH
(i.e., for 40 MSPS: Newt
S_E
and t
H_DR
+ t
H_DR
+ t
S_DR
H_DR
/2
ENC
/2.
ENC
) is calculated relative to 65 MSPS (50% duty cycle) and is dependent on t
V
Analog Input Current25mA
Digital Input Voltage0AV
CC
V
Digital Output Current4mA
ENVIRONMENTAL
2
I100% production tested.
II100% production tested at 25°C, and guaranteed by
design and characterization at temperature extremes.
III Sample tested only.
IV Parameter is guaranteed by design and characterization
testing.
VParameter is a typical value only.
Operating Temperature Range
(Ambient)–25+85°C
Maximum Junction Temperature150°C
Lead Temperature (Soldering, 10 sec)300°C
Storage Temperature Range (Ambient)–65+150 °C
NOTES
1
Absolute maximum ratings are limiting values to be applied individually, and
beyond which the serviceability of the circuit may be impaired. Functional
operability is not necessarily implied. Exposure to absolute maximum rating
conditions for an extended period of time may affect device reliability.
2
Typical thermal impedances (52-terminal LQFP); θJA = 33°C/W; θJC = 11°C/W.
These measurements were taken on a 6 layer board in still air with a solid ground
plane.
AD6644AST-40–25°C to +85°C (Ambient)52-Terminal LQFP (Low-Profile Quad Plastic Flatpack)ST-52
AD6644AST-65–25°C to +85°C (Ambient)52-Terminal LQFP (Low-Profile Quad Plastic Flatpack)ST-52
AD6644ST/PCBEvaluation Board with AD6644AST–65
CAUTION
ESD (electrostatic discharge) sensitive device. Electrostatic charges as high as 4000 V readily
accumulate on the human body and test equipment and can discharge without detection. Although
the AD6644 features proprietary ESD protection circuitry, permanent damage may occur on devices
subjected to high-energy electrostatic discharges. Therefore, proper ESD precautions are recommended to avoid performance degradation or loss of functionality.
11AINAnalog Input.
12AINComplement of AIN; Differential Analog Input.
20C1Internal Voltage Reference; bypass to ground with 0.1 µF microwave
24C2Internal Voltage Reference; bypass to ground with 0.1 µF microwave
31DNCDo not connect this pin.
32OVROverrange Bit; high indicates analog input exceeds ± FS.
35DMIDOutput Data Voltage Midpoint; approximately equal to (DVCC)/2.
36D0 (LSB)Digital Output Bit (Least Significant Bit); Two’s Complement
37–41, 44–50D1–D5, D6–D12Digital Output Bits in Two’s Complement.
51D13 (MSB)Digital Output Bit (Most Significant Bit); Two’s Complement.
52DRYData Ready Output.
3.3 V Power Supply (Digital) Output Stage Only.
2.4 V (Analog Reference). Bypass to ground with 0.1 µF microwave
chip capacitor.
5 V Analog Power Supply.
chip capacitor.
chip capacitor.
DV
GND
V
REF
GND
ENCODE
ENCODE
GND
AV
AV
GND
AIN
AIN
GND
CC
CC
CC
PIN CONFIGURATION
D8
D9
D10
D11
D12
D13 (MSB)
DRY
1
PIN 1
IDENTIFIER
2
3
4
5
6
7
8
9
10
11
12
13
14 151617 18
CC
AV
GND
AD6644
TOP VIEW
(Not to Scale)
19 20 21 22 23 24 25 26
CC
CC
AV
DNC = DO NOT CONNECT
GND
AV
GND
C1
D7
GND
D6
AV
CC
D4
D5
GND
DV
40414243444546474849505152
39
D3
D2
38
37
D1
D0 (LSB)
36
DMID
35
34
GND
DV
33
CC
32
OVR
31
DNC
AV
30
CC
29
GND
AV
28
CC
GND
27
GND
CC
C2
GND
AV
CC
–6–
REV. 0
Page 93
AD6644
DEFINITIONS OF SPECIFICATIONS
Analog Bandwidth
The analog input frequency at which the spectral power of the
fundamental frequency (as determined by the FFT analysis) is
reduced by 3 dB.
Aperture Delay
The delay between the 50% point of the rising edge of the
ENCODE command and the instant at which the analog input
is sampled.
Aperture Uncertainty (Jitter)
The sample-to-sample variation in aperture delay.
Differential Analog Input Resistance, Differential Analog Input
Capacitance, and Differential Analog Input Impedance
The real and complex impedances measured at each analog
input port. The resistance is measured statically and the capacitance and differential input impedances are measured with a
network analyzer.
Differential Analog Input Voltage Range
The peak-to-peak differential voltage that must be applied to the
converter to generate a full-scale response. Peak differential voltage
is computed by observing the voltage on a single pin and subtracting the voltage from the other pin, which is 180 degrees out
of phase. Peak-to-peak differential is computed by rotating the
inputs phase 180 degrees and taking the peak measurement again.
The difference is then computed between both peak measurements.
Differential Nonlinearity
The deviation of any code width from an ideal 1 LSB step.
Encode Pulsewidth/Duty Cycle
Pulsewidth high is the minimum amount of time that the
ENCODE pulse should be left in Logic 1 state to achieve rated
performance; pulsewidth low is the minimum time ENCODE
pulse should be left in low state. See timing implications of
changing t
in text. At a given clock rate, these specs define
ENCH
an acceptable ENCODE duty cycle.
Full-Scale Input Power
Expressed in dBm. Computed using the following equation:
2
Power
Harmonic Distortion, 2nd
Full Scale
=
10
log
V
Full Scalerms
Z
||
.
0 001
Input
The ratio of the rms signal amplitude to the rms value of the
second harmonic component, reported in dBc.
Harmonic Distortion, 3rd
The ratio of the rms signal amplitude to the rms value of the
third harmonic component, reported in dBc.
Integral Nonlinearity
The deviation of the transfer function from a reference line
measured in fractions of 1 LSB using a “best straight line”
determined by a least-square curve fit.
Minimum Conversion Rate
The encode rate at which the SNR of the lowest analog signal
frequency drops by no more than 3 dB below the guaranteed limit.
Maximum Conversion Rate
The encode rate at which parametric testing is performed.
Output Propagation Delay
The delay between a differential crossing of ENCODE and
ENCODE and the time when all output data bits are within
valid logic levels.
Noise (For Any Range Within the ADC)
VZ
=××
NOISE
|| .–0 001 10
FSSignal
dBmdBFS
10
Where Z is the input impedance, FS is the full scale of the device
for the frequency in question, SNR is the value for the particular
input level and Signal is the signal level within the ADC reported
in dB below full scale. This value includes both thermal and
quantization noise.
Power Supply Rejection Ratio
The ratio of a change in input offset voltage to a change in power
supply voltage.
Signal-to-Noise-and-Distortion (SINAD)
The ratio of the rms signal amplitude (set 1 dB below full scale)
to the rms value of the sum of all other spectral components,
including harmonics, but excluding dc.
Signal-to-Noise Ratio (Without Harmonics)
The ratio of the rms signal amplitude (set at 1 dB below full scale)
to the rms value of the sum of all other spectral components,
excluding the first five harmonics and dc.
Spurious-Free Dynamic Range (SFDR)
The ratio of the rms signal amplitude to the rms value of the peak
spurious spectral component. The peak spurious component may
or may not be a harmonic. May be reported in dBc (i.e., degrades
as signal level is lowered), or dBFS (always related back to converter full scale).
Two-Tone Intermodulation Distortion Rejection
The ratio of the rms value of either input tone to the rms value
of the worst third order intermodulation product; reported in dBc.
Two-Tone SFDR
The ratio of the rms value of either input tone to the rms value
of the peak spurious component. The peak spurious component
may or may not be an IMD product. May be reported in dBc
(i.e., degrades as signal level is lowered), or in dBFS (always
related back to converter full scale).
Worst Other Spur
The ratio of the rms signal amplitude to the rms value of the worst
spurious component (excluding the 2nd and 3rd harmonic)
reported in dBc.
Figure 12. Harmonics vs. Analog Frequency (Nyquist)
0
ENCODE = 65MSPS
–10
AIN = 30MHz @ –1dBFS
SNR = 73.5dB
–20
SFDR = 85dBc
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
5 101520
0
FREQUENCY – MHz
Figure 10. Single Tone at 30 MHz
REV. 0
2530
–9–
75
74
73
72
PHASE NOISE OF ANALOG SOURCE
71
SNR – dB
DEGRADES PERFORMANCE
70
69
68
1020507080
0
LOW NOISE ANALOG SOURCE
AIN = –1dBFS
ENCODE = 65MSPS
30406090
ANALOG FREQUENCY – MHz
Figure 13. Noise vs. Analog Frequency (IF)
100
Page 96
AD6644
100
95
WORST OTHER SPUR
90
85
80
75
70
HARMONICS – dBc
65
60
55
0
HARMONICS (2nd, 3rd)
1020507080
30406090
ANALOG FREQUENCY – MHz
ENCODE = 65MSPS
AIN = –1dBFS
100
Figure 14. Harmonics vs. Analog Frequency (IF)
120
110
100
ENCODE = 65MSPS
90
AIN = 15.5MHz
80
70
60
50
40
30
20
10
WORST-CASE SPURIOUS – dBFS and dBc
0
–70
–800
–60–50–30
ANALOG INPUT POWER LEVEL – dBFS
–40
dBFS
dBc
SFDR = 90dB
REFERENCE LINE
–20–10
Figure 15. Single Tone SFDR
0
–10
–20
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
5 101520
0
FREQUENCY – MHz
ENCODE = 65MSPS
AIN = 15MHz,
15.5MHz @ –7dBFS
NO DITHER
2530
Figure 17. Two Tones at 15 MHz and 15.5 MHz
110
100
ENCODE = 65MSPS
90
F1 = 15MHz
80
F2 = 15.5MHz
70
60
50
40
30
20
10
WORST-CASE SPURIOUS – dBFS and dBc
0
–67–27–17
–77
dBc
SFDR = 90dB
REFERENCE LINE
–57–47–37
INPUT POWER LEVEL – (F1 = F2) dBFS
dBFS
–7
Figure 18. Two-Tone SFDR
0
ENCODE = 65MSPS
–10
AIN = 19MHz,
19.5MHz @ –7dBFS
–20
NO DITHER
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
5 101520
0
FREQUENCY – MHz
Figure 16. Two Tones at 19 MHz and 19.5 MHz
2530
–10–
100
95
90
85
80
75
70
65
SNR, WORST SPURIOUS – dB and dBc
60
0
WORST SPUR
106070
204050
308090
ENCODE FREQUENCY – MHz
AIN = 2.2MHz @ –1dBFS
SNR
Figure 19. SNR, Worst Spurious vs. Encode
REV. 0
Page 97
AD6644
0
ENCODE = 65MSPS
–10
AIN = 15.5MHz @ –29.5dBFS
NO DITHER
–20
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
05101520
FREQUENCY – MHz
2530
Figure 20. 1M FFT Without Dither
100
ENCODE = 65MSPS
90
AIN = 15.5MHz
NO DITHER
80
70
60
50
40
30
20
WORST-CASE SPURIOUS – dBc
10
0
–900
–80–30–10
–60–20
–70–50 –40
ANALOG INPUT POWER LEVEL – dBFS
SFDR = 90dB
REFERENCE LINE
Figure 21. SFDR Without Dither
0
ENCODE = 65MSPS
–10
AIN = 15.5MHz @ –29.5dBFS
DITHER @ –19dBm
–20
–30
–40
–50
–60
–70
–80
–90
–100
–110
–120
–130
5 101520
0
Figure 23. 1M FFT with Dither
100
ENCODE = 65MSPS
90
AIN = 15.5MHz
DITHER = –19dBm
80
70
60
50
40
30
20
WORST-CASE SPURIOUS – dBc
10
0
–80–30–10
–90
Figure 24. SFDR with Dither
FREQUENCY – MHz
SFDR = 90dB
REFERENCE LINE
–60–20
–70–50 –40
ANALOG INPUT POWER LEVEL – dBFS
2530
SFDR = 100dB
REFERENCE LINE
0
95
90
WORST SPUR
85
80
75
70
SNR, WORST SPURIOUS – dB and dBc
65
–15.0
30.5MHz
–10.0
ENCODE INPUT POWER – dBm
2.2MHz
2.2MHz
30.5MHz
–5.010.0
0
Figure 22. SNR, Worst Spurious vs. Clamped Encode
Power (See Figure 25)
REV. 0
ENCODE = 65MSPS
SNR
5.0
15.0
–11–
Page 98
AD6644
THEORY OF OPERATION
The AD6644 analog-to-digital converter (ADC) employs a three
stage subrange architecture. This design approach achieves the
required accuracy and speed while maintaining low power and
small die size.
As shown in the functional block diagram, the AD6644 has
complementary analog input pins, AIN and AIN . Each analog
input is centered at 2.4 V and should swing ± 0.55 V around
this reference (Figure 2). Since AIN and AIN are 180 degrees
out of phase, the differential analog input signal is 2.2 V peakto-peak.
Both analog inputs are buffered prior to the first track-and-hold,
TH1. The high state of the ENCODE pulse places TH1 in hold
mode. The held value of TH1 is applied to the input of a 5-bit
coarse ADC1. The digital output of ADC1 drives a 5-bit digitalto-analog converter, DAC1. DAC1 requires 14 bits of precision
which is achieved through laser trimming. The output of DAC1
is subtracted from the delayed analog signal at the input of TH3
to generate a first residue signal. TH2 provides an analog pipeline delay to compensate for the digital delay of ADC1.
The first residue signal is applied to a second conversion stage
consisting of a 5-bit ADC2, 5-bit DAC2, and pipeline TH4.
The second DAC requires 10 bits of precision which is met by
the process with no trim. The input to TH5 is a second residue
signal generated by subtracting the quantized output of DAC2
from the first residue signal held by TH4. TH5 drives a final
6-bit ADC3.
The digital outputs from ADC1, ADC2, and ADC3 are added
together and corrected in the digital error correction logic to
generate the final output data. The result is a 14-bit parallel
digital CMOS-compatible word, coded as two's complement.
APPLYING THE AD6644
Encoding the AD6644
The AD6644 encode signal must be a high quality, extremely low
phase noise source to prevent degradation of performance. Maintaining 14-bit accuracy places a premium on encode clock phase
noise. SNR performance can easily degrade by 3 dB to 4 dB
with 70 MHz input signals when using a high-jitter clock source.
See Analog Devices’ Application Note AN-501, “Aperture Uncertainty and ADC System Performance” for complete details.
For optimum performance, the AD6644 must be clocked
differentially. The encode signal is usually ac-coupled into the
ENCODE and ENCODE pins via a transformer or capacitors.
These pins are biased internally and require no additional bias.
Shown below is one preferred method for clocking the AD6644.
The clock source (low jitter) is converted from single-ended to
differential using an RF transformer. The back-to-back Schottky
diodes across the transformer secondary limit clock excursions
into the AD6644 to approximately 0.8 V p-p differential. This
helps prevent the large voltage swings of the clock from feeding
through to the other portions of the AD6644, and limits the noise
presented to the ENCODE inputs. A crystal clock oscillator can
also be used to drive the RF transformer if an appropriate limiting
resistor (typically 100 Ω) is placed in the series with the primary.
If a low jitter ECL/PECL clock is available, another option is to
ac-couple a differential ECL/PECL signal to the encode input pins
as shown below. A device that offers excellent jitter performance
is the MC100LVEL16 (or same family) from Motorola.
VT
ECL/
PECL
0.1F
0.1F
VT
ENCODE
AD6644
ENCODE
Figure 26. Differential ECL for Encode
Analog Input
As with most new high-speed, high dynamic range analog-todigital converters, the analog input to the AD6644 is differential.
Differential inputs allow much improvement in performance
on-chip as signals are processed through the analog stages. Most
of the improvement is a result of differential analog stages having
high rejection of even order harmonics. There are also benefits
at the PCB level. First, differential inputs have high commonmode rejection to stray signals such as ground and power noise.
Also, they provide good rejection to common-mode signals such as
local oscillator feedthrough.
The AD6644 input voltage range is offset from ground by 2.4 V.
Each analog input connects through a 500 Ω resistor to a 2.4 V
bias voltage and to the input of a differential buffer (Figure 2). The
resistor network on the input properly biases the followers for maximum linearity and range. Therefore, the analog source driving the
AD6644 should be ac-coupled to the input pins. Since the differential input impedance of the AD6644 is 1 kΩ, the analog input
power requirement is only –2 dBm, simplifying the driver amplifier
in many cases. To take full advantage of this high-input impedance, a 20:1 transformer would be required. This is a large ratio
and could result in unsatisfactory performance. In this case, a
lower step-up ratio could be used. The recommended method for
driving the analog input of the AD6644 is to use a 4:1 RF transformer. For example, if R
were set to 60.4 Ω and RS were set
T
to 25 Ω, along with a 4:1 transformer, the input would match
to a 50 Ω source with a full-scale drive of 4.8 dBm. Series resistors (R
) on the secondary side of the transformer should be
S
used to isolate the transformer from A/D. This will limit the
amount of dynamic current from the A/D flowing back into
the secondary of the transformer. The terminating resistor (RT)
should be placed on the primary side of the transformer.
R
ANALOG INPUT
SIGNAL
T1–4T
R
T
S
R
S
0.1F
AIN
AD6644
AIN
–12–
Figure 27. Transformer-Coupled Analog Input Circuit
REV. 0
Page 99
AD6644
JITTER – ps
0
SNR – dB
0.1
55
0.20.30.40.50.6
60
65
70
75
80
AIN = 190MHz
AIN = 150MHz
AIN = 110MHz
AIN = 30MHz
AIN = 70MHz
In applications where dc-coupling is required, a new differential
output op amp from Analog Devices, the AD8138, can be used
to drive the AD6644 (Figure 28). The AD8138 op amp provides
single-ended-to-differential conversion, which reduces overall
system cost and minimizes layout requirements.
C
F
499⍀
5V
25⍀
AD8138
25⍀
499⍀
C
V
0.1F
IN
499⍀
499⍀
V
OCM
Figure 28. DC-Coupled Analog Input Circuit
Power Supplies
Care should be taken when selecting a power source. Linear
supplies are strongly recommended. Switching supplies tend to
have radiated components that may be “received” by the AD6644.
Each of the power supply pins should be decoupled as closely to
the package as possible using 0.1 µF chip capacitors.
The AD6644 has separate digital and analog power supply pins.
The analog supplies are denoted AV
pins are denoted DV
. AVCC and DV
CC
CC
power supplies. This is because the fast digital output swings
can couple switching current back into the analog supplies. Note
that AV
fied for DV
must be held within 5% of 5 V. The AD6644 is speci-
CC
= 3.3 V as this is a common supply for digital ASICs.
CC
Output Loading
Care must be taken when designing the data receivers for the
AD6644. It is recommended that the digital outputs drive a
series resistor (e.g. 100 Ω) followed by a gate like 74LCX574.
To minimize capacitive loading, there should only be one gate
on each output pin. An example of this is shown in the evaluation
board schematic shown in Figure 30. The digital outputs of the
AD6644 have a constant output slew rate of 1 V/ns. A typical
CMOS gate combined with a PCB trace will have a load of
approximately 10 pF. Therefore, as each bit switches, 10 mA
(10 pF⫻ 1 V ⫼ 1 ns) of dynamic current per bit will flow in or out
of the device. A full scale transition can cause up to 140 mA
(14 bits ⫻ 10 mA/bit) of current to flow through the output stages.
The series resistors should be placed as close to the AD6644 as
possible to limit the amount of current that can flow into the output stage. These switching currents are confined between ground
and the DV
pin. Standard TTL gates should be avoided since
CC
they can appreciably add to the dynamic switching currents of
the AD6644. It should also be noted that extra capacitive loading
will increase output timing and invalidate timing specifications.
Digital output timing is guaranteed with 10 pF loads.
Layout Information
The schematic of the evaluation board (Figure 30) represents a
typical implementation of the AD6644. A multilayer board is
recommended to achieve the best results. It is highly recommended that high-quality, ceramic chip capacitors be used to
decouple each supply pin to ground directly at the device. The
pinout of the AD6644 facilitates ease of use in the implementa-
REV. 0
AIN
AD6644
AIN
V
REF
DIGITAL
OUTPUTS
and the digital supply
should be separate
CC
tion of high frequency, high resolution design practices. All of
the digital outputs are segregated to two sides of the chip, with
the inputs on the opposite side for isolation purposes.
Care should be taken when routing the digital output traces.
To prevent coupling through the digital outputs into the analog
portion of the AD6644, minimal capacitive loading should be
placed on these outputs. It is recommended that a fan-out of
only one gate be used for all AD6644 digital outputs.
The layout of the Encode circuit is equally critical. Any noise
received on this circuitry will result in corruption in the digitization process and lower overall performance. The Encode clock
must be isolated from the digital outputs and the analog inputs.
Jitter Considerations
The signal-to-noise ratio (SNR) for an ADC can be predicted.
When normalized to ADC codes, Equation 1 accurately predicts
the SNR based on three terms. These are jitter, average DNL
error, and thermal noise. Each of these terms contributes to the
noise within the converter.
2
()
+
1
ε
–log
SNRft
=×
f
ANALOG
t
J RMS
20
()
+×××+
2
π
N
2
ANALOGRMS
= analog input frequency.
= rms jitter of the encode (rms sum of encode
V
2
J
NOISE RMS
N
2
2
(1)
12
/
source and internal encode circuitry).
ε= average DNL of the ADC (typically 0.41 LSB).
N= Number of bits in the ADC.
V
NOISE RMS
= V rms thermal noise referred to the analog input
of the ADC (typically 2.5 LSB).
For a 14-bit analog-to-digital converter like the AD6644, aperture jitter can greatly affect the SNR performance as the analog
frequency is increased. The chart below shows a family of curves
that demonstrates the expected SNR performance of the AD6644
as jitter increases. The chart is derived from the above equation.
For a complete discussion of aperture jitter, please consult
Analog Devices’ Application Note AN-501, “Aperture
Uncertainty and ADC System Performance.”
Figure 29. SNR vs. Jitter
–13–
Page 100
AD6644
EVALUATION BOARD
The evaluation board for the AD6644 is straightforward,
containing all required circuitry for evaluating the device. The
only external connections required are power supplies, clock,
and the analog inputs. The evaluation board includes the option
for an onboard clock oscillator for ENCODE.
Power to the analog supply pins of the AD6644 is connected via
the power terminal block (PCTB2). Power for the digital interface
is supplied via pin 1 of J6. The J2 connector mates directly with
SoftCell
boards, allowing complete evaluation of system performance.
The analog input is connected via a BNC connector AIN, which
is transformer-coupled to the AD6644 inputs. The transformer
has a turns ratio of 1:4 to reduce the amount of input power
required to drive the AD6644.
The Encode signal may be generated using an onboard crystal
oscillator, U5. The on-board oscillator may be replaced by an
external encode source via the SMA connector labeled OPT_CLK
or BNC connector labeled ENCODE. If an external source is
used, it must be a high-quality and very low-phase noise source.
The AD6644 output data is latched using 74LCX574 (U7, U2)
latches. The clock for these latches is determined by selecting
jumper E3–E4 or E4–E5. E3 to E5 is a just a gate delayed version of the clock, while connecting E4 to E5 utilizes the Data
Ready of the AD6644 to latch the output data. A clock is also
distributed with the output data (J2) that is labeled BUFLAT
(Pin 19 and 20, J2).
DC-Coupling Only)
Coupling Only)
–14–
REV. 0
Loading...
+ hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.