INTERSIL CA3304AM

CA3304, CA3304A
®
November 2002
T
ODUC
DUC T
TE PR
E
E P RO a t
L
T
O
U
S
IT
T
OB
S
ter
B
LE SU Support Cen /tsc
IB
S
S
O
l
m
P
a
o
FOR A our Technic ww.intersil.c
t
c
w
a
r
t
o
n
IL
co
NT E RS
1-888-I
4-Bit, 25 MSPS,
Flash A/D Converters
Features
Description
•
•
•
•
The Intersil CA3304 is a CMOS parallel (FLASH) analog-todigital converter designed for applications demanding both
low-power consumption and high speed digitization. Digitizing at 25MHz, for example, requires only about 35mW.
•
•
•
•
•
CMOS/SOS Low Power with Video Speed (Typ) . . 25mW
Parallel Conversion Technique
Single Power Supply Voltage . . . . . . . . . . . 3V to 7.5V
25MHz Sampling Rate (40ns Conversion Time) at 5V
Supply
4-Bit Latched Three-State Output with Overflow and
Data Change Outputs
1/ LSB Maximum Nonlinearity (A Version)
8
Inherent Resistance to Latch-Up Due to SOS Process
Bipolar Input Range with Optional Second Supply
Wide Input Bandwidth (Typ) . . . . . . . . . . . . . . . . 25MHz
The CA3304 operates over a wide, full-scale signal input
voltage range of 0.5V up to the supply voltage. Power
consumption is as low as 10mW, depending upon the clock
frequency selected.
The intrinsic high conversion rate makes the CA3304 types
ideally suited for digitizing high speed signals. The overflow
bit makes possible the connection of two or more CA3304s
in series to increase the resolution of the conversion system.
A series connection of two CA3304s may be used to produce a 5-bit, 25MHz converter. Operation of two CA3304s in
parallel doubles the conversion speed (i.e., increases the
sampling rate from 25MHz to 50MHz). A data change pin
indicates when the present output differs from the previous,
thus allowing compaction of data storage.
Applications
•
•
•
•
•
•
•
•
•
•
High Speed A/D Conversion
Ultrasound Signature Analysis
Transient Signal Analysis
High Energy Physics Research
General-Purpose Hybrid ADCs
Optical Character Recognition
Radar Pulse Analysis
Motion Signature Analysis
Robot Vision
RSSI Circuits
Sixteen paralleled auto-balanced voltage comparators measure the input voltage with respect to a known reference to
produce the parallel-bit outputs in the CA3304. Fifteen comparators are required to quantize all input voltage levels in this
4-bit converter, and the additional comparator is required for
the overflow bit.
Part Number Information
PART NUMBER LINEARITY (INL, DNL)
SAMPLING RATE
TEMP. RANGE (oC)
25MHz (40ns)
-40 to 85
PACKAGE
PKG. NO.
CA3304E
±0.25 LSB
CA3304AE
±0.125 LSB
25MHz (40ns)
-40 to 85
16 Ld PDIP
E16.3
CA3304M
±0.25 LSB
25MHz (40ns)
-40 to 85
16 Ld SOIC (W)
M16.3
CA3304AM
±0.125 LSB
25MHZ (40ns)
-40 to 85
16 Ld SOIC (W)
M16.3
CA3304D
±0.25 LSB
25MHz (40ns)
-55 to 125
16 Ld SBDIP
D16.3
CA3304AD
±0.125 LSB
25MHz (40ns)
-55 to 125
16 Ld SBDIP
D16.3
16 Ld PDIP
E16.3
Pinout
CA3304 (SBDIP, PDIP, SOIC)
TOP VIEW
BIT 1 (LSB) 1
16 VDD
BIT 2 2
15 CLK
BIT 3 3
14 VAA-
BIT 4 4
13 VREF -
DATA CHANGE (DC) 5
12 VREF +
OVERFLOW (OF) 6
11 VIN
CE2 7
10 VAA+
VSS 8
9 CE1
CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures.
1-888-INTERSIL or 321-724-7143 | Intersil (and design) is a registered trademark of Intersil Americas Inc.
Copyright © Intersil Americas Inc. 2002. All Rights Reserved
All other trademarks mentioned are the property of their respective owners.
1
FN1790.3
CA3304, CA3304A
Absolute Maximum Ratings
Thermal Information
DC Supply Voltage Range (VDD or VAA+)
(Voltage Referenced to VSS or VAA- Terminal,
Whichever is More Negative) . . . . . . . . . . . . . . . . . . -0.5V to +8V
Input Voltage Range
CE1, CE2 Inputs . . . . . . . . . . . . . . . . . . . VSS -0.5V to VDD +0.5V
Clock, VREF+, VREF-, VIN Inputs . . . . . VAA- -0.5V to VAA- +0.5V
DC Input Current, Any Input . . . . . . . . . . . . . . . . . . . . . . . . . ±20mA
Thermal Resistance (Typical, Note 1)
θJA (oC/W) θJC (oC/W)
SBDIP Package. . . . . . . . . . . . . . . . . . . .
80
22
PDIP Package . . . . . . . . . . . . . . . . . . . . .
90
N/A
SOIC Package. . . . . . . . . . . . . . . . . . . . .
100
N/A
Maximum Junction Temperature
Ceramic Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175oC
Plastic Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150oC
Maximum Storage Temperature Range (TSTG) . . . -65oC to 150oC
Maximum Lead Temperature (Soldering 10s). . . . . . . . . . . . . 300oC
(SOIC - Lead Tips Only)
Operating Conditions
Recommended Supply Voltage Range (VDD or VAA+). . . . . 3V to 7.5V
Recommended VAA+ Voltage Range. . . . . . VDD -1V to VDD +2.5V
Recommended VAA- Voltage Range . . . . . . . VSS -2.5V to VSS +1V
Operating Temperature
CA3304D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -55oC to 125oC
CA3304E, CA3304M. . . . . . . . . . . . . . . . . . . . . . . . -40oC to 85oC
CAUTION: Stresses above those listed in “Absolute Maximum Ratings” may cause permanent damage to the device. This is a stress only rating and operation
of the device at these or any other conditions above those indicated in the operational sections of this specification is not implied.
NOTE:
1. θJA is measured with the component mounted on an evaluation PC board in free air.
Electrical Specifications
TA = 25oC, VREF+ = 2V, VDD = VAA+ = 5V, VAA- = VREF - = VSS = GND, fCLK = 25MHz
Unless Otherwise Specified
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNITS
SYSTEM PERFORMANCE
Resolution
Input Errors
4
-
-
Bits
Integral Linearity
Error
CA3304A
-
±0.1
±0.125
LSB
CA3304
-
±0.125
±0.25
LSB
Differential Linearity
Error
CA3304A
-
±0.1
±0.125
LSB
CA3304
-
±0.125
±0.25
LSB
Offset Error
(Unadjusted)
CA3304A
-
-
±0.75
LSB
CA3304
-
-
±1.0
LSB
Gain Error
(Unadjusted)
CA3304A
-
-
±0.75
LSB
CA3304
-
-
±1.0
LSB
-
3
-
ns
DYNAMIC CHARACTERISTICS (Input Signal Level 0.5dB Below Full Scale)
Conversion Timing
Aperture Delay
Signal to Noise Ratio, SNR
RMS Signal
=
RMS Noise
fS = 25MHz, fIN = 100kHz
-
23.7
-
dB
fS = 25MHz, fIN = 5MHz
-
23.6
-
dB
Signal to Noise Ratio, SINAD
RMS Signal
=
RMS Noise + Distortion
fS = 25MHz, fIN = 100kHz
-
23.4
-
dB
fS = 25MHz, fIN = 5MHz
-
22.8
-
dB
Total Harmonic Distortion, THD
fS = 25MHz, fIN = 100kHz
-
-34.5
-
dBc
fS = 25MHz, fIN = 5MHz
-
-31.0
-
dBc
fS = 25MHz, fIN = 100kHz
-
3.67
-
Bits
fS = 25MHz, fIN = 5MHz
-
3.57
-
Bits
0.5
-
VAA
V
-
10
-
pF
-
150
200
µA
Effective Number of Bits, ENOB
ANALOG INPUTS
Input Range
Full Scale Input Range
Input Loading
Input Capacitance
Input Current
(Notes 1, 4)
VIN = 2V (Note 2)
2
CA3304, CA3304A
Electrical Specifications
TA = 25oC, VREF+ = 2V, VDD = VAA+ = 5V, VAA- = VREF - = VSS = GND, fCLK = 25MHz
Unless Otherwise Specified (Continued)
PARAMETER
TEST CONDITIONS
Allowable Input Bandwidth
(Note 4)
-3dB Input Bandwidth
MIN
TYP
MAX
UNITS
-
25
fCLK/2
MHz
-
40
-
MHz
REFERENCE INPUTS
Input Range
Input Loading
VREF+ Range
(Note 4)
VAA+0.5
-
VAA+
V
VREF- Range
(Note 4)
VAA-
-
VAA+ 0.5
V
Resistor Ladder Impedance
VIN = 5V, CLK = Low
640
-
960
Ω
Maximum VIN, Low
CLOCK
(Notes 3, 4)
-
-
0.3 x VAA
V
CE1, CE2
(Note 4)
-
-
0.3 x
VDD
V
CLOCK
(Notes 3, 4)
0.7 x VAA
-
-
V
CE1, CE2
(Note 4)
0.7 x
VDD
-
-
V
DIGITAL INPUTS
Digital Input
Minimum VIN, High
Input Leakage, Except CLK
V = 0V, 5V
-
-
±1
µA
Input Leakage, CLK
(Note 3)
-
±100
±150
µA
DIGITAL OUTPUTS
Digital Outputs
Output Low (Sink) Current
VO = 0.4V
6
-
-
mA
Output High (Source) Current
VO = 4.6V
-3
-
-
mA
Three-State Leakage Current
VO = 0V, 5V
-
±0.2
±5
µA
25
35
-
MSPS
20
-
-
ns
TIMING CHARACTERISTICS
Conversion Timing
Maximum Conversion Speed
CLK = Square Wave
Auto-Balance Time (φ1)
Sample Time (φ2)
Output Timing
20
-
5000
ns
Data Valid Delay
(Note 4)
-
30
40
ns
Data Hold Time
(Note 4)
15
25
-
ns
Output Enable Time
-
15
-
ns
Output Disable Time
-
10
-
ns
Continuous Clock
-
5.5
-
mA
Continuous φ2
-
0.4
-
mA
POWER SUPPLY CHARACTERISTICS
Device Current, IAA
Device Current, IDD
VAA+ = 5V,
VSS = CE1 = VAA- = CLK = GND
VAA+ = 7V
Continuous φ1
-
2
-
mA
Continuous Clock
-
1.5
-
mA
Continuous φ2
-
5
10
mA
Continuous φ1
-
5
20
mA
NOTES:
1. Full scale input range, VREF + - VREF -, may be in the range of 0.5V to VAA+ -VAA- volts. Linearity errors increase at lower full scale ranges,
however.
2. Input current is due to energy transferred to the input at the start of the sample period. The average value is dependent on input and VDD
voltage.
3. The CLK input is a CMOS inverter with a 50kΩ feedback resistor. It operates from the VAA+ and VAA- supplies. It may be AC-coupled
with a 1VP-P minimum source.
4. Parameter not tested, but guaranteed by design or characterization.
3
CA3304, CA3304A
Pin Descriptions
PIN NUMBER
NAME
1
Bit 1
Bit 1 (LSB).
2
Bit 2
Bit 2.
3
Bit 3
Bit 3.
4
Bit 4
Bit 4 (MSB).
5
DC
Data Change.
6
OF
Overflow.
7
CE2
Three-State Output Enable Input, active low. See the Chip Enable Truth Table.
8
VSS
Digital Ground.
Three-State Output Enable Input, active high. See the Chip Enable Truth Table.
9
CE1
10
VAA+
11
VIN
DESCRIPTION
Output Data Bits
(High = True)
Analog Power Supply, +5V.
Analog Signal Input.
12
VREF+
Reference Voltage Positive Input.
13
VREF-
Reference Voltage Negative Input.
14
VAA-
Analog Ground.
15
CLK
Clock Input.
16
VDD
Digital Power Supply, +5V.
CHIP ENABLE TRUTH TABLE
CE1
CE2
BIT 1 - BIT 4
DC, OF
0
1
Valid
Valid
1
1
Three-State
Valid
X
0
Three-State
Three-State
X = Don't Care
TABLE 1. OUTPUT CODE TABLE
INPUT VOLTAGE (V)
CODE
DESCRIPTION
VREF + = 1V
VREF - = -1V
1.6V
0V
2V
0V
OUTPUT CODE
3.2V
0V
4.8V
0V
OF
B4
B3
B2
B1
DECIMAL
COUNT
Zero
-1.000
0
0
0
0
0
0
0
0
0
0
1 LSB
-0.875
0.1
0.125
0.2
0.3
0
0
0
0
1
1
2 LSB
-0.750
0.2
0.250
0.4
0.6
0
0
0
1
0
2
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
1/ Full Scale -1 LSB
2
1/ Full Scale
2
1/ Full Scale +1 LSB
2
-0.125
0.7
0.875
1.4
2.1
0
0
1
1
1
7
0
0.8
1.000
1.6
2.4
0
1
0
0
0
8
0.125
0.9
1.125
1.8
2.7
0
1
0
0
1
9
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Full Scale -1 LSB
0.750
1.4
1.750
2.8
4.2
0
1
1
1
0
14
Full Scale
0.875
1.5
1.875
3.0
4.5
0
1
1
1
1
15
Overflow
1.000
1.6
2.000
3.2
4.8
1
1
1
1
1
31
Step Size
0.125
0.1
0.125
0.2
0.3
NOTE:
1. The voltages listed are the ideal centers of each output code shown as a function of its associated reference voltage See Ideal Transfer
Curve Figure 6. The output code should exist for an input equal to the ideal center voltage ±1/2 of the step size.
4
CA3304, CA3304A
Functional Diagram
φ2
φ1
φ2 φ1
φ1
φ1
VAA+
VDD
10
16
THREE-STATE
DRIVERS
OUTPUT
REGISTER
5 DATA
CHANGE
D Q
CLK
VIN
11
D
1/ R
2
12
LATCH
16
†CAB #16
VREF +
COUNT
16
Q
R
COUNT
ENCODER
8
LOGIC
Q
D
ARRAY
LATCH
8
R
†CAB #8
D Q
CLK
6 OVERFLOW
D Q
CLK
4 BIT 4
D Q
CLK
3 BIT 3
D Q
CLK
2 BIT 2
D Q
CLK
1 BIT 1 (LSB)
R
R
VREF - 1
D
/2R
13
LATCH
0
†CAB COMPARATOR #1
50kΩ
COUNT
1
Q
φ1 (AUTO BALANCE)
CLOCK
15
9 CE1
φ2 (SAMPLE UNKNOWN)
14
8
VAA-
VSS
7 CE2
†Cascaded Auto Balance (CAB)
NOTE: CE1 and CE2 inputs and data outputs have standard CMOS protection networks to VDD and VSS . Analog inputs and clock have
standard CMOS protection networks to VAA+ and VAA-.
Timing Diagrams
DATA SHIFTED INTO
OUTPUT REGISTERS
1
CLOCK
φ1
AUTO
BALANCE
0
COMPARATOR DATA
LATCHED
φ2
SAMPLE 1
1
B1 - B4, DC & OF
AUTO
BALANCE
SAMPLE 2
DATA VALID 0
0
AUTO
BALANCE
SAMPLE 3
DATA VALID 1
DATA VALID 2
tHO
tD
FIGURE 1. TIMING DIAGRAM
CE1
CE2
tDIS
BITS 1-4
tEN
HIGH
IMPEDANCE
tDIS
tEN
HIGH
IMPEDANCE
HIGH
DC, OF
IMPEDANCE
FIGURE 2. OUTPUT ENABLE/DISABLE TIMING
5
CA3304, CA3304A
Timing Diagrams
(Continued)
SAMPLE ENDS
φ2
CLOCK
SAMPLE ENDS
φ1
φ2
φ1
CLOCK
φ1
φ2
tD
OUTPUT
φ1
φ2
tD
OLD DATA
NEW DATA
OUTPUT
FIGURE 3A.
With φ2 as standby state (fastest method, but standby limited to 5µs
maximum)
OLD DATA + 1
OLD DATA
NEW DATA
FIGURE 3B.
With φ1 as standby state (indefinite standby, double pulse needed)
SAMPLE ENDS
φ1
φ2
CLOCK
φ1
φ2
φ2
tD
OUTPUT
INVALID DATA
OLD DATA
NEW DATA
FIGURE 3C.
With φ2 as standby state (indefinite standby, lower power than 3B)
FIGURE 3. PULSE-MODE TIMING DIAGRAMS
Typical Performance Curves
8
40
7
38
6
IDD + IAA (MA)
TD (ns)
36
34
32
5
4
30
3
28
-50
-25
0
25
50
75
100
2
TEMPERATURE (oC)
5
10
15
20
25
30
fS (MHz)
FIGURE 4. DATA DELAY vs TEMPERATURE
FIGURE 5. DEVICE CURRENT vs SAMPLE FREQUENCY
6
CA3304, CA3304A
Typical Performance Curves
(Continued)
0.10
0.25
0.09
0.22
0.08
0.17
NON-LINEARITY (LSB)
NON-LINEARITY (LSB)
0.20
INL
0.15
0.12
DNL
0.10
0.07
0.06
0.04
0.03
0.02
0.02
0.01
0
10 20 30 40 50
TEMPERATURE (oC)
60
70
80
0.00
90
INL
0.05
0.05
0.00
-40 -30 -20 -10
DNL
1
2
3
4
5
REFERENCE VOLTAGE (V)
FIGURE 6. NON-LINEARITY vs TEMPERATURE
FIGURE 7. NON-LINEARITY vs REFERENCE VOLTAGE
0.50
4.00
0.45
3.80
0.40
3.60
0.35
3.40
INL
0.30
ENOB (LSB)
NON-LINEARITY (LSB)
0.07
0.25
0.20
0.15
0.10
3.20
3.00
2.80
2.60
2.40
0.05
2.20
DNL
0.00
15
20
25
fS (MHz)
30
2.00
-40
35
FIGURE 8. NON-LINEARITY vs SAMPLE FREQUENCY
-30 -20 -10
0
10 20 30 40 50
TEMPERATURE (oC)
60
70
FIGURE 9. EFFECTIVE BITS vs TEMPERATURE
7
80 90
CA3304, CA3304A
Typical Performance Curves
(Continued)
7.00
4.00
6.80
3.80
6.60
3.60
6.40
IDD (mA)
ENOB (LSB)
3.40
3.20
3.00
6.20
6.00
2.80
5.80
2.60
5.60
2.40
5.40
2.20
5.20
5.00
-40
2.00
0
1
2
3
4
5
6
7
8
9
10
-30 -20 -10
0
10
20
30
40
50
60
70
80
TEMPERATURE (oC)
fI (MHz)
FIGURE 10. EFFECTIVE BITS vs INPUT FREQUENCY
FIGURE 11. DEVICE CURRENT vs TEMPERATURE
27Ω
+5V SUPPLY
CA3304
+
4.7µF TAN
2V REFERENCE
+
4.7µF TAN
0.1µF
CER
VAA+
VDD
CE2
VREF+
0.1µF
CER
REMOTE
2V INTO 50Ω
SOURCE
DC, OF,
B1-B4
VIN
CLK
VREF- CE1
VSS
VAA-
50Ω
ANALOG
GROUND
0.1µF
CER
+
4.7µF TAN
OUTPUT DATA
CMOS CLOCK
SOURCE
DIGITAL
GROUND
FIGURE 12A. TYPICAL CA3304 UNIPOLAR CIRCUIT CONFIGURATION
8
90
CA3304, CA3304A
Typical Performance Curves
(Continued)
27Ω
+5V SUPPLY
CA3304
+
4.7µF TAN
+1V REFERENCE
REMOTE
±1V INTO 50Ω
SOURCE
0.1µF
CER
0.1µF
CER
VAA+
VDD
CE2
0.1µF
CER
VREF +
+
DC, OF,
VIN B1 - B4
-1V
REFERENCE
50Ω
OUTPUT DATA
IN914
CMOS CLOCK
SOURCE
VREF - CLK
0.1µF
CER
VAA-
0.1µF
CER
CE1
VSS
4.7µF TAN
10K
0.001µF
-1.5V SUPPLY
4.7µF TAN
ANALOG
GROUND
DIGITAL
GROUND
FIGURE 12B. TYPICAL CA3304 BIPOLAR CIRCUIT CONFIGURATION
FIGURE 12.
9
CA3304, CA3304A
Description
Continuous Clock Operation
Device Operation
One complete conversion cycle can be traced through the
CA3304 via the following steps. (Refer to timing diagram
Figure 3). The rising edge of the clock input will start a
“sample” phase. During this entire “High” state of the clock,
the 16 comparators will track the input voltage and the 16
latches will track the comparator outputs. At the falling edge
of the clock, all 16 comparator outputs are captured by the
16 latches. This ends the “sample” phase and starts the
“auto balance” phase for the comparators. During this “Low”
state of the clock the output of the latches propagates
through the decode array and a 6-bit code appears at the D
inputs of the output registers. On the next rising edge of the
clock, this 6-bit code is shifted into the output registers and
appears with time delay tD as valid data at the output of the
three-state drivers. This also marks the start of a new
“sample” phase, thereby repeating the conversion process
for this next cycle.
A sequential parallel technique is used by the CA3304
converter to obtain its high speed operation. The sequence
consists of the “Auto Balance” phase and the “Sample
Unknown” phase (Refer to the circuit diagram). Each
conversion takes one clock cycle (see Note). The “Auto
Balance” (φ1) occurs during the Low period of the clock
cycle, and the “Sample Unknown” (φ2) occurs during the
High period of the clock cycle.
NOTE: This device requires only a single-phase clock. The terminology
of φ1 and φ2 refers to the High and Low periods of the same clock.
During the “Auto Balance” phase, a transmission-gate switch
is used to connect each of 16 commutating capacitors to
their associated ladder reference tap. Those tap voltages will
be as follows:
VTAP(N) = [(VREF/16) x N] - [VREF/(2 x 16)]
Pulse Mode Operation
= VREF [(2N - 1)/32],
For sampling high speed nonrecurrent or transient data, the
converter may be operated in a pulse mode in one of three
ways. The fastest method is to keep the converter in the
Sample Unknown phase, φ2, during the standby state. The
device can now be pulsed through the Auto Balance phase
with as little as 20ns. The analog value is captured on the
leading edge of φ1 and is transferred into the output registers
on the trailing edge of φ1. We are now back in the standby
state, φ2, and another conversion can be started within
20ns, but not later than 5µs due to the eventual droop of the
commutating capacitors. Another advantage of this method
is that it has the potential of having the lowest power drain.
The larger the time ratio between φ2 and φ1, the lower the
power consumption. (See Timing Diagram Figure 3A).
Where:
VTAP(N) = Reference ladder tap voltage at point N,
VREF = Voltage across VREF - to VREF +, and
N = Tap number (1 through 16).
The other side of the capacitor is connected to a singlestage inverting amplifier whose output is shorted to its input
by a switch. This biases the amplifier at its intrinsic trip point,
which is approximately (VDD - VSS)/2. The capacitors now
charge to their associated tap voltages, priming the circuit
for the next phase.
In the “Sample Unknown” phase, all ladder tap switches are
opened, the comparator amplifiers are no longer shorted,
and VIN is switched to all 16 capacitors. Since the other end
of the capacitor is now looking into an effectively open circuit, any voltage that differs from the previous tap voltage
will appear as a voltage shift at the comparator amplifiers. All
comparators whose tap voltages were lower than VIN will
drive the comparator outputs to a “low” state. All comparators whose tap voltages were higher than VIN will drive the
comparator outputs to a “high” state. A second, capacitorcoupled, auto-zeroed amplifier further amplifies the outputs.
The second method uses the Auto Balance phase, φ1, as
the standby state. In this state the converter can stay
indefinitely waiting to start a conversion. A conversion is
performed by strobing the clock input with two φ2 pulses.
The first pulse starts a Sample Unknown phase and
captures the analog value in the comparator latches on the
trailing edge. A second φ2 pulse is needed to transfer the
date into the output registers. This occurs on the leading
edge of the second pulse. The conversion now takes place
in 40ns, but the repetition rate may be as slow as desired.
The disadvantage to this method is the slightly higher device
dissipation due to the low ratio of φ2 to φ1. (See Timing
Diagram Figure 3B).
The status of all these comparator amplifiers are stored at the
end of this phase (φ2), by a secondary latching amplifier stage.
Once latched, the status of the 16 comparators is decoded by
a 16 to 5 bit decode array and the results are clocked into a
storage register at the rising edge of the next φ2.
For applications requiring both indefinite standby and lowest
power, standby can be in the φ2 (Sample Unknown) state
with two φ1 pulses to generate valid data (see Figure 3C).
The conversion process now takes 60ns. [Note that the
above numbers do not include the tD (Output Delay) time.]
If the input is greater than 31/32 x VREF , the overflow output
will go “high”. (The bit outputs will remain high). If the output
differs from that of the previous conversion, the data change
output will go “high”.
Increased Accuracy
A three-state buffer is used at the output of the 7 storage
registers which are controlled by two chip-enable signals.
CE1 will independently disable B1 through B4 when it is in a
high state. CE2 will independently disable B1 through B4
and the OF and DC buffers when it is in the low state.
In most case the accuracy of the CA3304 should be
sufficient without any adjustments. In applications where
accuracy is of utmost importance, two adjustments can be
made to obtain better accuracy; i.e., offset trim and gain trim.
10
CA3304, CA3304A
Offset Trim
Digital Input And Output Levels
In general offset correction can be done in the preamp
circuitry by introducing a DC shift to VIN or by the offset trim
of the op amp. When this is not possible the VREF - input can
be adjusted to produce an offset trim.
The clock input is a CMOS inverter operating from and with
logic input levels determined by the VAA supplies. If VAA+ or
VAA- are outside the range of the digital supplies, it may be
necessary to level shift the clock input to meet the required
30% to 70% of VAA input swing. Figure 12B shows an example for a negative VAA-.
The theoretical input voltage to produce the first transition is
1/ LSB. The equation is as follows:
2
An alternate way of driving the clock is to capacitively couple
the pin from a source of at least 1VP-P . An internal 50kΩ
feedback resistor will keep the DC level at the intrinsic trip
point. Extremely non-symmetrical clock waveforms should
be avoided, however.
VIN (0 to 1 transition) = 1/2 LSB = 1/2(VREF/16)
= VREF/32.
Adjust offset by applying this input voltage and adjusting the
VREF - voltage or input amplifier offset until an output code
alternating between 0 and 1 occurs.
The remaining digital inputs and outputs are referenced to
VDD and VSS . If TTL or other lower voltage sources are to
drive the CA3304, either pull-up resistors or CD74HCT
series “QMOS” buffers are recommended.
Gain Trim
In general the gain trim can also be done in the preamp circuitry
by introducing a gain adjustment for the op amp. When this is
not possible, then a gain adjustment circuit should be made to
adjust the reference voltage. To perform this trim, VIN should
be set to the 15 to overflow transition. That voltage is 1/2 LSB
less than VREF + and is calculated as follows:
5-Bit Resolution
To obtain 5-bit resolution, two CA3304s can be wired
together. Necessary ingredients include an open-ended ladder network, an overflow indicator, three-state outputs, and
chip-enable controls - all of which are available on the
CA3304.
VlN (15 to 16 transition) = VREF - VREF/32
= VREF (31/32).
The first step for connecting a 5-bit circuit is to totem-pole
the ladder networks, as illustrated in Figure 13. Since the
absolute-resistance value of each ladder may vary, external
trim of the mid-reference voltage may be required.
To perform the gain trim, first do the offset trim and then
apply the required VIN for the 15 to overflow transition. Now
adjust VREF+ until that transition occurs on the outputs.
Layout, Input And Supply Considerations
The overflow output of the lower device now becomes the
fifth bit. When it goes high, all counts must come from the
upper device. When it goes low, all counts must come from
the lower device. This is done simply by connecting the
lower overflow signal to the CE1 control of the lower A/D
converter and the CE2 control of the upper A/D converter.
The three-state outputs of the two devices (bits 1 through 4)
are now connected in parallel to complete the circuitry.
The CA3304 should be mounted on a ground-planed,
printed-circuit board, with good high-frequency decoupling
capacitors mounted as close as possible. If the supply is
noisy, decouple VAA+ with a resistor as shown in Figure
12A. The CA3304 outputs current spikes to its input at the
start of the auto-balance and sample clock phases. A low
impedance source, such as a locally-terminated 50Ω coax
cable, should be used to drive the input terminal. A fastsettling buffer such as the HA-5033, HA-5242, or CA3450
should be used if the source is high impedance. The VREF
terminals also have current spikes, and should be well
bypassed.
Definitions
Dynamic Performance Definitions
Fast Fourier Transform (FFT) techniques are used to evaluate
the dynamic performance of the CA3304. A low distortion sine
wave is applied to the input, it is sampled, and the output is
stored in RAM. The data is then transformed into the frequency domain with a 4096 point FFT and analyzed to evaluate the dynamic performance of the A/D. The sine wave input
to the part is -0.5dB down from full scale for all these tests.
Care should be taken to keep digital signals away from the
analog input, and to keep digital ground currents away from
the analog ground. If possible, the analog ground should be
connected to digital ground only at the CA3304.
Bipolar Operation
The CA3304, with separate analog (VAA+, VAA-) and digital
(VDD , VSS) supply pins, allows true bipolar or negative input
operation. The VAA- pin may be returned to a negative
supply (observing maximum voltage ratings to VAA+ or VDD
and recommended rating to VSS), thus allowing the VREFpotential also to be negative. Figure 12B shows operation
with an input range of -1V to +1V. Similarly, VAA+ and
VREF + could be maintained at a higher voltage than VDD ,
for an input range above the digital supply.
Signal-to-Noise (SNR)
SNR is the measured RMS signal to RMS noise at a specified input and sampling frequency. The noise is the RMS
sum of all of the spectral components except the fundamental and the first five harmonics.
Signal-to-Noise + Distortion Ratio (SINAD)
SINAD is the measured RMS signal to RMS sum of all other
spectral components below the Nyquist frequency excluding DC.
11
Effective Number of Bits (ENOB)
+5V
The effective number of bits (ENOB) is derived from the
SINAD data. ENOB is calculated from:
+FULL
SCALE
REF.
ENOB = (SINAD - 1.76 + VCORR)/6.02,
where: VCORR = 0.5dB.
Total Harmonic Distortion (THD)
THD is the ratio of the RMS sum of the first 5 harmonic
components to the RMS value of the measured input signal.
BUFFER
INPUT
1K
ADJUST
CENTER
Operating and Handling Considerations
+5V
HANDLING
All inputs and outputs of CMOS devices have a network for
electrostatic protection during handling. Recommended handling practices for CMOS devices are described in
lCAN-6525. “Guide to Better Handling and Operation of
CMOS Integrated Circuits.”
VAA+
VDD
DC
NC
OF
NC
VREF +
VIN
VREF VAA-
B4
B3
B2
B1
VSS
CE1
CLK
CE2
CLOCK
INPUT
CA3304
NC
CLK
VAA+
DC
VDD
B4
B4
VREF +
B3
VIN
B2
B3
B2
B5 MSB
OF
VREF B1
VAACE1
VSS
CE2
B1
+5V
CA3304
OPERATING
FIGURE 13. TYPICAL CA3304 5-BIT CONFIGURATION
Operating Voltage
During operation near the maximum supply voltage limit, care
should be taken to avoid or suppress power supply turn-on
and turn-off transients, power supply ripple, or ground noise;
any of these conditions must not cause the power supply
voltages to exceed the absolute maximum rating.
OVERFLOW
DECIMAL COUNT
Input Signals
To prevent damage to the input protection circuit, input signals
should never be greater than VDD or VAA+ nor less than VSS
or VAA- (depending upon which supply the protection network
is referenced. See Maximum Ratings.). Input currents must
not exceed 20mA even when the power supply is off.
Unused Inputs
A connection must be provided at every input terminal. All
unused input terminals must be connected to either VDD or
VSS , whichever is appropriate.
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
V0 V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12 V13 V14 V15 V16
Output Short Circuits
INPUT VOLTAGE
Shorting of outputs to any supply potential may damage
CMOS devices by exceeding the maximum device dissipation.
FIGURE 14. IDEAL TRANSFER CURVE
All Intersil semiconductor products are manufactured, assembled and tested under ISO9000 quality systems certification.
Intersil products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design and/or specifications at any time without notice.
Accordingly, the reader is cautioned to verify that data sheets are current before placing orders. Information furnished by Intersil is believed to be accurate and reliable. However, no responsibility is assumed by Intersil or its subsidiaries 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 Intersil or its subsidiaries.
For information regarding Intersil Corporation and its products, see web site http://www.intersil.com
12