INTERSIL CA3306CE

CA3306, CA3306A,
CA3306C
UCT
P RO D E P RO DU CT
E
T
E
L
O BS O
r at
STITUT
E SUB upport Cente sc
L
IB
S
POS
al S
om/t
FOR A our Technic ww.intersil.c
t
w
c
conta TERSIL or
IN
1-888®
November 2002
6-Bit, 15 MSPS,
Flash A/D Converters
Features
Description
•
•
•
•
•
•
The CA3306 family are CMOS parallel (FLASH) analog-to-digital
converters designed for applications demanding both low power
consumption and high speed digitization. Digitizing at 15MHz, for
example, requires only about 50mW.
CMOS Low Power with Video Speed (Typ) . . . . . 70mW
Parallel Conversion Technique
Signal Power Supply Voltage . . . . . . . . . . . 3V to 7.5V
15MHz Sampling Rate with Single 5V Supply
6-Bit Latched Three-State Output with Overflow Bit
Pin-for-Pin Retrofit for the CA3300
The CA3306 family operates over a wide, full scale signal input voltage range of 1V up to the supply voltage. Power consumption is as
low as 15mW, depending upon the clock frequency selected. The
CA3306 types may be directly retrofitted into CA3300 sockets, offering improved linearity at a lower reference voltage and high operating speed with a 5V supply.
Applications
•
•
•
•
•
•
•
•
•
•
TV Video Digitizing
Ultrasound Signature Analysis
Transient Signal Analysis
High Energy Physics Research
High Speed Oscilloscope Storage/Display
General Purpose Hybrid ADCs
Optical Character Recognition
Radar Pulse Analysis
Motion Signature Analysis
Robot Vision
The intrinsic high conversion rate makes the CA3306 types ideally
suited for digitizing high speed signals. The overflow bit makes possible the connection of two or more CA3306s in series to increase
the resolution of the conversion system. A series connection of two
CA3306s may be used to produce a 7-bit high speed converter.
Operation of two CA3306s in parallel doubles the conversion speed
(i.e., increases the sampling rate from 15MHz to 30MHz).
Sixty-four paralleled auto balanced comparators measure the input
voltage with respect to a known reference to produce the parallel bit
outputs in the CA3306. Sixty-three comparators are required to
quantize all input voltage levels in this 6-bit converter, and the additional comparator is required for the overflow bit.
Part Number Information
PART NUMBER LINEARITY (INL, DNL)
TEMP. RANGE (oC)
SAMPLING RATE
PACKAGE
PKG. NO.
CA3306E
±0.5 LSB
15MHz (67ns)
-40 to 85
18 Ld PDIP
E18.3
CA3306CE
±0.5 LSB
10MHz (100ns)
-40 to 85
18 Ld PDIP
E18.3
CA3306M
±0.5 LSB
15MHz (67ns)
-40 to 85
20 Ld SOIC
M20.3
CA3306CM
±0.5 LSB
10MHz (100ns)
-40 to 85
20 Ld SOIC
M20.3
CA3306D
±0.5 LSB
15MHz (67ns)
-55 to 125
18 Ld SBDIP
D18.3
CA3306CD
±0.5 LSB
10MHz (100ns)
-55 to 125
18 Ld SBDIP
D18.3
CA3306J3
±0.5 LSB
15MHz (67ns)
-55 to 125
20 Ld CLCC
J20.B
CA3306J3
±0.5 LSB
10MHz (100ns)
-55 to 125
20 Ld CLCC
J20.B
Pinouts
CE2 6
CLK 7
PHASE 8
VREF + 9
14 B2
13 B1 (LSB)
12 VDD
11 VIN
10 VREF -
B4
16 B2
2
B5
VZ 5
3
1 20 19
VZ 5
18 REF
CENTER
17 B3
NC 6
16 B2
VSS 4
CE2 6
15 B1 (LSB)
CE1 7
14 VDD
CLK 8
13 NC
9 10 11 12 13
PHASE 9
12 VIN
VREF + 10
11 VREF -
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
15 B1 (LSB)
CE2 7
14 VDD
CE1 8
VIN
CE2 5
NC 4
19 B4
18 REF
CENTER
17 B3
VSS 3
VREF -
VZ 4
20 B5
VREF +
VSS 3
17 B4
REF
16
CENTER
15 B3
(MSB) B6 1
OVERFLOW 2
PHASE
OVERFLOW 2
18 B5
CA3306 (CLCC)
TOP VIEW
OVERFLOW
B6
(MSB)
NC
(MSB) B6 1
CA3306 (SOIC)
TOP VIEW
CLK
CA3306 (PDIP, SBDIP)
TOP VIEW
FN3102.2
CA3306, CA3306A, CA3306C
Functional Block Diagram
VIN
φ1
φ1
φ1
φ2
R/2
COMP
64
φ2
VREF+
THREE-STATE
R
COMP
63
R
R
≅ 120Ω
REF
CENTER
COMP
32
R
COMPARATOR
LATCHES
AND
ENCODER
LOGIC
R
COMP
2
R
D Q
CL
OVERFLOW
D Q
CL
B6 (MSB)
D Q
CL
B5
D Q
CL
B4
D Q
CL
B3
D Q
CL
B2
D Q
CL
B1 (LSB)
VREFCOMP
1
R/2
≅ 50kΩ
CLOCK
CE1
φ2 (SAMPLE UNKNOWN)
PHASE
φ1 (AUTO BALANCE)
CE2
ZENER
6.2V NOMINAL
DIODE
VDD
VSS
VSS
Typical Application Circuit
OF
B6
1
B6
2
OF
B4 17
3
VSS
RC 16
B2
4
VZ
B3 15
B1
(LSB)
5
CE2
B2 14
6
CE1
B1 13
7
CLK
VDD 12
8
PH
9
VREF+
CA3306
B5 18
B5
0.1µF
6.2V
560Ω
+12V
B4
B3
DATA
OUTPUT
+5V
+5V
CLOCK
+12V
0.2µF
5kΩ
+
CA741CE
VIN 11
0.1µF
2
VREF- 10
SIGNAL
INPUT
10µF
CA3306, CA3306A, CA3306C
Absolute Maximum Ratings
Thermal Information
DC Supply Voltage Range, VDD
Voltage Referenced to VSS Terminal . . . . . . . . . . . -0.5V to +8.5V
Input Voltage Range
All Inputs Except Zener. . . . . . . . . . . . . . . . . -0.5V to VDD + 0.5V
DC Input Current
CLK, PH, CE1, CE2, VIN . . . . . . . . . . . . . . . . . . . . . . . . . ±20mA
Thermal Resistance (Typical, Note 1)
θJA (oC/W) θJC (oC/W)
SBDIP Package. . . . . . . . . . . . . . . . . . . .
75
24
PDIP Package . . . . . . . . . . . . . . . . . . . . .
95
N/A
SOIC Package. . . . . . . . . . . . . . . . . . . . .
115
N/A
CLCC Package . . . . . . . . . . . . . . . . . . . .
80
28
Maximum Junction Temperature
Hermetic Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175oC
Plastic Packages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150oC
Maximum Storage Temperature Range . . . . . . . . . . -65oC to 150oC
Maximum Lead Temperature (Soldering 10s). . . . . . . . . . . . . 300oC
(SOIC - Lead Tips Only)
Operating Conditions
Supply Voltage Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3V to 8V
Temperature Range (TA)
Ceramic Package (D Suffix) . . . . . . . . . . . . . . . . . -55oC to 125oC
Plastic Package (E or M Suffix) . . . . . . . . . . . . . . . -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, VDD = 5V, VREF + = 4.8V, VSS = VREF - = GND, Clock = 15MHz Square Wave for CA3306
or CA3306A, 10MHz for CA3306C
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNITS
6
-
-
Bits
CA3306, CA3306C
-
±0.25
±0.5
LSB
CA3306A
-
±0.2
±0.25
LSB
CA3306, CA3306C
-
±0.25
±0.5
LSB
CA3306A
-
±0.2
±0.25
LSB
SYSTEM PERFORMANCE
Resolution
Integral Linearity Error, INL
Differential Linearity Error,
DNL
CA3306, CA3306C (Note 1)
-
±0.5
±1
LSB
CA3306A
-
±0.25
±0.5
LSB
CA3306, CA3306C (Note 2)
-
±0.5
±1
LSB
CA3306A
-
±0.25
±0.5
LSB
Gain Temperature Coefficient
-
+0.1
-
mV/oC
Offset Temperature Coefficient
-
-0.1
-
mV/oC
10
13
-
MSPS
15
Offset Error (Unadjusted)
Gain Error (Unadjusted)
DYNAMIC CHARACTERISTICS (Input Signal Level 0.5dB Below Full Scale)
Maximum Conversion Speed
CA3306C
CA3306, CA3306A
20
-
MSPS
(Note 4)
φ1, φ2 ≥ Minimum
12
-
-
MSPS
18
-
-
MSPS
(Note 4)
DC
-
fCLOCK/2
MHz
-
30
-
MHz
fS = 15MHz, fIN = 100kHz
-
34.6
-
dB
fS = 15MHz, fIN = 5MHz
-
33.4
-
dB
fS = 15MHz, fIN = 100kHz
-
34.2
-
dB
fS = 15MHz, fIN = 5MHz
-
29.0
-
dB
Total Harmonic Distortion, THD
fS = 15MHz, fIN = 100kHz
-
-46.0
-
dBc
fS = 15MHz, fIN = 5MHz
-
-30.0
-
dBc
Effective Number of Bits, ENOB
fS = 15MHz, fIN = 100kHz
-
5.5
-
Bits
fS = 15MHz, fIN = 5MHz
-
4.5
-
Bits
Maximum Conversion Speed
CA3306C
CA3306, CA3306A
Allowable Input Bandwidth
-3dB Input Bandwidth
Signal to Noise Ratio, SNR
RMSSignal
= -------------------------------RMSNoise
Signal to Noise Ratio, SINAD
RMSSignal
= -----------------------------------------------------------RMSNoise+Distortion
3
CA3306, CA3306A, CA3306C
Electrical Specifications
TA = 25oC, VDD = 5V, VREF + = 4.8V, VSS = VREF - = GND, Clock = 15MHz Square Wave for CA3306
or CA3306A, 10MHz for CA3306C (Continued)
PARAMETER
TEST CONDITIONS
MIN
TYP
MAX
UNITS
4.8
VDD +
0.5
V
ANALOG INPUTS
Positive Full Scale Input Range
(Notes 3, 4)
1
Negative Full Scale Input Range
(Notes 3, 4)
-0.5
0
VDD - 1
V
-
15
-
pF
-
-
±500
µA
5.4
6.2
7.4
V
-
12
25
Ω
-
-0.5
-
mV/oC
650
1100
1550
Ω
Input Capacitance
Input Current
VIN = 4.92V, VDD = 5V
INTERNAL VOLTAGE REFERENCE
Zener Voltage
IZ = 10mA
Zener Dynamic Impedance
IZ = 10mA, 20mA
Zener Temperature Coefficient
REFERENCE INPUTS
Resistor Ladder Impedance
DIGITAL INPUTS
Maximum VIN , Logic 0
All Digital Inputs (Note 4)
-
-
0.3 x
VDD
V
Maximum VIN , Logic 1
All Digital Inputs (Note 4)
0.7 x
VDD
-
-
V
Digital Input Current
Except CLK, VIN = 0V, 5V
-
±1
±5
µA
Digital Input Current
CLK Only
-
±100
±200
µA
VOUT = 0V, 5V
-
±1
±5
µA
DIGITAL OUTPUTS
Digital Output Three-State Leakage
Digital Output Source Current
VOUT = 4.6V
-1.6
-
-
mA
Digital Output Sink Current
VOUT = 0.4V
3.2
-
-
mA
50
-
∞
ns
33
-
∞
33
-
5000
TIMING CHARACTERISTICS
Auto Balance Time (φ1)
CA3306C
Sample Time (φ2)
CA3306C
CA3306, CA3306A
(Note 4)
CA3306, CA3306A
ns
22
-
5000
ns
Aperture Delay
-
8
-
ns
Aperture Jitter
-
100
-
psP-P
CA3306C
-
35
50
ns
CA3306, CA3306A
-
30
40
ns
15
25
-
ns
Output Enable Time, tEN
-
20
-
ns
Output Disable Time, tDIS
-
15
-
ns
-
11
20
mA
-
14
25
mA
-
7.5
15
mA
Output Data Valid Delay, tD
Output Data Hold Time, tH
(Note 4)
POWER SUPPLY CHARACTERISTICS
IDD Current, Refer to Figure 4 CA3306C
Continuous Conversion (Note 4)
CA3306, CA3306A
IDD Current
Continuous φ1
NOTES:
1. OFFSET ERROR is the difference between the input voltage that causes the 00 to 01 output code transition and (VREF + - VREF -)/128.
2. GAIN ERROR is the difference the input voltage that causes the 3F16 to overflow output code transition and (VREF + - VREF -) x 127/128.
3. The total input voltage range, set by VREF + and VREF -, may be in the range of 1 to (VDD + 1) V.
4. Parameter not tested, but guaranteed by design or characterization.
4
CA3306, CA3306A, CA3306C
Timing Waveforms
COMPARATOR DATA IS LATCHED
CLOCK IF
PHASE IS HIGH
DECODED DATA IS SHIFTED TO OUTPUT REGISTERS
φ2
CLOCK IF
PHASE IS LOW
φ1
φ2
φ1
φ2
AUTO
BALANCE
SAMPLE
N+1
AUTO
BALANCE
SAMPLE
N+2
tD
tH
DATA
N-2
DATA
N-1
DATA
N
FIGURE 1. INPUT-TO-OUTPUT
CE1
CE2
tDIS
tEN
tDIS
HIGH
IMPEDANCE
BITS 1-6
tDIS
HIGH
IMPEDANCE
DATA
DATA
DATA
HIGH
IMPEDANCE
OF
DATA
DATA
FIGURE 2. OUTPUT ENABLE
5
CA3306, CA3306A, CA3306C
Timing Waveforms
SAMPLE ENDS
SAMPLE ENDS
φ1
φ2
CLOCK
φ2
CLOCK
φ1
φ2
φ1
φ2
tD
tD
NEW DATA
OLD DATA
OUTPUT
φ1
OLD
DATA
OUTPUT
OLD
DATA +1
NEW
DATA
FIGURE 3B.
FIGURE 3A.
SAMPLE ENDS
CLOCK
φ2
φ1
φ2
φ1
φ2
tD
OUTPUT
INVALID
DATA
OLD DATA
NEW
DATA
FIGURE 3C.
FIGURE 3. PULSE MODE
Typical Performance Curves
125
50
AMBIENT TEMPERATURE (oC)
TA = 25oC, VREF + = VDD
VIN = 0 TO VREF + SINE WAVE AT fCLK/2
40
IDD (mA)
DISSIPATION LIMITED
30
20
VDD
VDD
VDD
VDD
= 8V
= 7V
= 6V
= 5V
10
fCLK = 1MHz
fCLK = 3MHz
fCLK = 10MHz
fCLK = 15MHz
fCLK = 20MHz
100
MAXIMUM AMBIENT
TEMPERATURE - PLASTIC
75
50
VREF + = VDD
VIN = 0 TO VREF + SINE WAVE AT fCLK/2
VDD = 3V
ZENER NOT CONNECTED
25
0.1
1
CLOCK FREQUENCY (MHz)
3
10
4
5
6
7
8
VDD (V)
FIGURE 4. TYPICAL IDD AS A FUNCTION OF VDD
FIGURE 5. TYPICAL MAXIMUM AMBIENT TEMPERATURE AS
A FUNCTION OF SUPPLY VOLTAGE
6
CA3306, CA3306A, CA3306C
Typical Performance Curves
(Continued)
0.35
TA = 25oC, VDD = 5V
fCLK = 15MHz
1.2
0.25
INTEGRAL
NON-LINEARITY (LSB)
NON-LINEARITY (LSB)
TA = 25oC, VREF = 4.8V
0.30 VDD = 5V
0.20
0.15
DIFFERENTIAL
0.10
0.05
0
0.1
0.8
INTEGRAL
0.6
0.4
DIFFERENTIAL
0.2
0
1
CLOCK FREQUENCY (MHz)
10
0
FIGURE 6. TYPICAL NON-LINEARITY AS A FUNCTION OF
CLOCK SPEED
2
3
REFERENCE VOLTAGE (V)
4
fCLK = 15MHz, VREF + = VDD
VREF - = VSS
INPUT CURRENT (µA)
+10
1
FIGURE 7. TYPICAL NON-LINEARITY AS A FUNCTION OF
REFERENCE VOLTAGE
VREF + = VDD , VREF - = VSS
+15
PEAK INPUT CURRENT (mA)
1.0
VDD = 8V
+5
VDD = 5V
0
-5
-10
-15
+400
VDD = 8V
+200
VDD = 5V
+0
-200
-400
-600
-800
0
1
2
3
4
5
6
7
INPUT VOLTAGE (V)
8
0
FIGURE 8. TYPICAL PEAK INPUT CURRENT AS A FUNCTION
OF INPUT VOLTAGE
1
2
3
4
5
6
7
INPUT VOLTAGE (V)
8
FIGURE 9. TYPICAL AVERAGE INPUT CURRENT AS A
FUNCTION OF INPUT VOLTAGE
7
5
CA3306, CA3306A, CA3306C
Typical Performance Curves
(Continued)
11
TA = 25oC, VREF+ = VDD
VIN = 0 TO VREF + SINE WAVE AT fCLK/2
10
30
9
20
IDD (mA)
DECODER
LIMITED
25
DISSIPATION
LIMITED
8
7
15
6
10
5
5
0
3
4
5
6
7
4
8
0
5
10
fS (MHz)
SUPPLY VOLTAGE (V)
FIGURE 10. TYPICAL MAXIMUM CLOCK FREQUENCY AS A
FUNCTION OF SUPPLY VOLTAGE
15
20
FIGURE 11. DEVICE CURRENT vs SAMPLE FREQUENCY
32.5
6.0
5.7
fS = 15MHz, fI = 1MHz
30.0
5.4
5.1
ENOB (LSB)
tD (ns)
27.5
25.0
22.5
4.8
4.5
4.2
3.9
20.0
3.6
17.5
3.3
15.0
-50
-25
0
25
50
75
3.0
-40 -30 -20 -10
100
TEMPERATURE (oC)
12.6
0.90
11.2
0.80
NON-LINEARITY (LSB)
1.00
9.8
IDD (mA)
8.4
7.0
5.6
4.2
30
40
50
60
70
80
90
0.60
70
80
90
0.40
0.30
1.4
0.10
70
80
0.0
-40 -30 -20 -10
90
INL
0.50
0.20
60
fS = 15MHz
0.70
2.8
10 20 30 40 50
TEMPERATURE (oC)
20
FIGURE 13. ENOB vs TEMPERATURE
14.0
0
10
TEMPERATURE (oC)
FIGURE 12. DATA DELAY vs TEMPERATURE
0.0
-40 -30 -20 -10
0
DNL
0
10
20
30
40
50
60
TEMPERATURE (oC)
FIGURE 14. IDD vs TEMPERATURE
FIGURE 15. NON-LINEARITY vs TEMPERATURE
8
CA3306, CA3306A, CA3306C
Typical Performance Curves
(Continued)
6.00
5.70
fS = 15MHz
5.40
ENOB (LSB)
5.10
4.80
4.50
4.20
3.90
3.60
3.30
3.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 4.50 5.00
fI (MHz)
FIGURE 16. ENOB vs INPUT FREQUENCY
Pin Descriptions
PIN NUMBER
DIP
SOIC
NAME
DESCRIPTION
1
1
B6
Bit 6, Output (MSB).
2
2
OF
Overflow, Output.
3
3, 4
VSS
Digital Ground.
4
5
VZ
Zener Reference Output.
5
6
CE2
Three-State Output Enable Input, Active Low. See Table 1.
6
7
CE1
Three-State Output Enable Input, Active High. See Table 1.
7
8
CLK
Clock Input.
8
9
Phase
Sample clock phase control input. When PHASE is low, “Sample Unknown” occurs
when the clock is low and “Auto Balance” occurs when the clock is high (see text).
9
10
VREF +
Reference Voltage Positive Input.
10
11
VREF -
Reference Voltage Negative Input.
11
12
VIN
Analog Signal Input.
12
13, 14
VDD
Power Supply, +5V.
13
15
B1
Bit 1, Output (LSB).
14
16
B2
Bit 2, Output.
15
17
B3
Bit 3, Output.
16
18
REF(CTR)
17
19
B4
Bit 4, Output.
18
20
B5
Bit 5, Output.
Reference Ladder Midpoint.
9
CA3306, CA3306A, CA3306C
TABLE 1. CHIP ENABLE TRUTH TABLE
CE1
CE2
B1 - B6
OF
0
1
Valid
Valid
1
1
Three-State
Valid
X
0
Three-State
Three-State
X = Don’t care
TABLE 2. OUTPUT CODE TABLE
(NOTE 1)
INPUT VOLTAGE
BINARY OUTPUT CODE (LSB)
CODE DESCRIPTION
VREF
6.40
(V)
VREF
5.12
(V)
VREF
4.80
(V)
VREF
3.20
(V)
OF
B6
B5
B4
B3
B2
B1
DECIMAL
COUNT
Zero
1 LSB
2 LSB
0.00
0.10
0.20
0.00
0.08
0.16
0.00
0.075
0.15
0.00
0.05
0.10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
1
2
•
•
•
•
1/ Full Scale - 1 LSB
2
1/ Full Scale
2
1/ Full Scale + 1 LSB
2
•
•
•
•
3.10
3.20
3.30
2.48
2.56
2.64
•
•
•
•
Full Scale - 1 LSB
Full Scale
Overflow
•
•
•
•
2.325
2.40
2.475
1.55
1.60
1.65
0
0
0
0
1
1
1
0
0
•
•
•
•
6.20
6.30
6.40
4.96
5.04
5.12
1
0
0
•
•
•
•
1
0
0
1
0
0
1
0
1
•
•
•
•
4.65
4.725
4.80
3.10
3.15
3.20
0
0
1
1
1
1
1
1
1
1
1
1
•
•
•
•
1
1
1
1
1
1
0
1
1
NOTE:
1. The voltages listed above are the ideal centers of each output code shown as a function of its associated reference voltage.
10
31
32
33
62
63
127
CA3306, CA3306A, CA3306C
Device Operation
Continuous Clock Operation
A sequential parallel technique is used by the CA3306
converter to obtain its high speed operation. The sequence
consists of the “Auto Balance” phase φ1 and the “Sample
Unknown” phase φ2. (Refer to the circuit diagram.) Each
conversion takes one clock cycle (see Note). With the phase
control low, the “Auto Balance” (φ1) occurs during the High
period of the clock cycle, and the “Sample Unknown” (φ2)
occurs during the low period of the clock cycle.
One complete conversion cycle can be traced through the
CA3306 via the following steps. (Refer to timing diagram,
Figure 1.) With the phase control in a “High” state, the rising
edge of the clock input will start a “sample” phase. During this
entire “High” state of the clock, the 64 comparators will track
the input voltage and the 64 latches will track the comparator
outputs. At the falling edge of the clock, after the specified
aperture delay, all 64 comparator outputs are captured by the
64 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 7-bit code appears at the D inputs of the
output registers. On the next rising edge of the clock, this 7bit code is shifted into the output registers and appears with
time delay to 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.
During the “Auto Balance” phase, a transmission-gate switch
is used to connect each of 64 commutating capacitors to
their associated ladder reference tap. Those tap voltages will
be as follows:
VTAP (N) = [(VREF/64) x N] - [VREF/(2 x 64)]
= VREF[(2N - 1)/126],
Where: VTAP (N) = reference ladder tap voltage at point N,
VREF = voltage across VREF - to VREF +,
N = tap number (I through 64).
Pulse Mode Operation
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 a single pulse. 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, 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 Waveform, Figure 3.)
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.
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 VlN is switched to all 64 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 VlN will
drive the comparator outputs to a “low” state. All comparators whose tap voltages were higher than VlN 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 data into the output registers. This occurs on the leading edge of the second
pulse. The conversion now takes slightly longer, but the repetition rate may be as slow as desired. The disadvantage to
this method is the higher device dissipation due to the low
ratio of φ2 to φ1. (See Timing Waveform, 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 64 comparators is decoded by
a 64-bit 7-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).
Valid data now appears two full clock cycles after starting the
conversion process.
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 81 through 86 when it is in a
high state. CE2 will independently disable B1 through B6
and the OF buffers when it is in the low state (Table 1).
Analog Input Considerations
The CA3306 input terminal is characterized by a small
capacitance (see Specifications) and a small voltagedependent current (See Typical Performance Curves). The
signal-source impedance should be kept low, however,
when operating the CA3306 at high clock rates.
To facilitate usage of this device a phase-control input is
provided which can effectively complement the clock as it
enters the chip. Also, an on-board zener is provided for use
as a reference voltage.
11
CA3306, CA3306A, CA3306C
If VIN for the first transition is greater than the theoretical,
then the 50Ω pot should be connected between VREF and a
negative voltage of about 2 LSBs. The trim procedure is as
stated previously.
The CA3306 outputs a short (less than 10ns) current spike
of up to several mA amplitude (See Typical Performance
Curves) at the beginning of the sample phase. (To a lesser
extent, a spike also appears at the beginning of auto balance.) The driving source must recover from the spike by the
end of the same phase, or a loss of accuracy will result.
Gain Trim
In general the gain trim can also be done in the preamp
circuitry by introducing a gain adjustment for the operational
amplifier. When this is not possible, then a gain adjustment
circuit should be made to adjust the reference voltage. To
perform this trim, VlN should be set to the 63 to overflow
transition. That voltage is 1/2 LSB less than VREF+ and is
calculated as follows:
A locally terminated 50Ω or 75Ω source is generally sufficient to drive the CA3306. If gain is required, a high speed,
fast settling operational amplifier, such as the HA-5033,
HA-2542, or HA5020 is recommended.
Digital Input And Output Interfacing
The two chip-enable and the phase-control inputs are standard CMOS units. They should be driven from less than 0.3
x VDD to at least 0.7 x VDD . This can be done from 74HC
series CMOS (QMOS), TTL with pull-up resistors, or, if VDD
is greater than the logic supply, open collector or open drain
drivers plus pull-ups. (See Figure 20.)
VlN (63 to 64 transition) = VREF - VREF/128
= VREF(127/128).
To perform the gain trim, first do the offset trim and then
apply the required VlN for the 63 to overtlow transition. Now
adjust VREF+ until that transition occurs on the outputs.
The clock input is more critical to timing variations, such as
φ1 becoming too short, for instance. Pull-up resistors should
generally be avoided in favor of active drivers. The clock
input may be capacitively coupled, as it has an internal 50kΩ
feedback resistor on the first buffer stage, and will seek its
own trip point. A clock source of at least 1VP-P is adequate,
but extremely non-symmetrical waveforms should be
avoided.
Midpoint Trim
The reference center (RC) is available to the user as the
midpoint of the resistor ladder. To trim the midpoint, the
offset and gain trims should be done first. The theoretical
transition from count 31 to 32 occurs at 311/2 LSBs. That
voltage is as follows:
VlN (31 to 32 transition) = 31.5 (VREF/64)
= VREF(63/128).
The output drivers have full rail-to-rail capability. If driving
CMOS systems with VDD below the VDD of the CA3306, a
CD74HC4050 or CD74HC4049 should be used to step down
the voltage. If driving LSTTL systems, no step-down should
be necessary, as most LSTTLs will take input swings up to
10V to 15V.
An adjustable voltage follower can be connected to the RC
pin or a 2K pot can be connected between VREF+ and
VREF- with the wiper connected to RC. Set VlN to the 31 to
32 transition voltage, then adjust the voltage follower or the
pot until the transition occurs on the output bits.
Although the output drivers are capable of handling typical
data bus loading, the capacitor charging currents will produce local ground disturbances. For this reason, an external
bus driver is recommended.
The Reference Center point can also be used to create
unique transfer functions. The user must remember, however,
that there is approximately 120Ω in series with the RC pin.
Increased Accuracy
Applications
In most cases the accuracy of the CA3306 should be
sufficient without any adjustments. In applications where
accuracy is of utmost importance, three adjustments can be
made to obtain better accuracy; i.e., offset trim, gain trim,
and midpoint trim.
7-Bit Resolution
To obtain 7-bit resolution, two CA3306s can be wired
together. Necessary ingredients include an open-ended ladder network, an overtlow indicator, three-state outputs, and
chip-enabler controls - all of which are available on the
CA3306.
Offset Trim
In general offset correction can be done in the preamp
circuitry by introducing a DC shift to VlN or by the offset trim
of the operational amplifier. When this is not possible the
VREF- input can be adjusted to produce an offset trim. The
theoretical input voltage to produce the first transition is 1/2
LSB. The equation is as follows:
The first step for connecting a 7-bit circuit is to totem-pole
the ladder networks, as illustrated in Figure 17. Since the
absolute resistance value of each ladder may vary, external
trim of the mid-reference voltage may be required.
The overflow output of the lower device now becomes the
seventh 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 6)
are now connected in parallel to complete the circuitry.
VIN (0 to 1 transition) = 1/2 LSB = 1/2(VREF/64)
= VREF/128.
If VlN for the first transition is less than the theoretical, then a
single-turn 50Ω pot connected between VREF- and ground will
accomplish the adjustment. Set VlN to 1/2 LSB and trim the pot
until the 0 to 1 transition occurs.
12
CA3306, CA3306A, CA3306C
Doubled Sampling Speed
Signal-to-Noise (SNR)
The phase control and both positive and negative true chip
enables allow the parallel connection of two CA3306s to
double the sampling speed. Figure 18 shows this configuration. One converter samples on the positive phase of the
clock, and the second on the negative. The outputs are also
alternately enabled. Care should be taken to provide a near
square-wave clock it operating at close to the maximum
clock speed for the devices.
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.
8-Bit to 12-Bit Conversion Techniques
Effective Number of Bits (ENOB)
To obtain 8-bit to 12-bit resolution and accuracy, use a feedforward conversion technique. Two A/D converters will be
needed to convert up to 11 bits; three A/D converters to convert 12 bits. The high speed of the CA3306 allows 12-bit
conversions in the 500ns to 900ns range.
The effective number of bits (ENOB) is derived from the
SINAD data. ENOB is calculated from:
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.
ENOB = (SINAD - 1.76 + VCORR)/6.02,
where:
The circuit diagram of a high-speed 12-bit A/D converter is
shown in Figure 19. In the feed-forward conversion method
two sequential conversions are made. Converter A first does
a coarse conversion to 6 bits. The output is applied to a 6-bit
D/A converter whose accuracy level is good to 12 bits. The
D/A converter output is then subtracted from the input voltage, multiplied by 32, and then converted by a second flash
A/D converter, which is connected in a 7-bit configuration.
The answers from the first and second conversions are
added together with bit 1 of the first conversion overlapping
bit 7 of the second conversion.
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.
Operating and Handling Considerations
HANDLING
All inputs and outputs of Intersil CMOS devices have a
network for electrostatic protection during handling. Recommended handling practices for CMOS devices are described
in AN6525. “Guide to Better Handling and Operation of
CMOS Integrated Circuits.”
When using this method, take care that:
• The linearity of the first converter is better than 1/2 LSB.
• An offset bias of 1 LSB (1/64) Is subtracted from the first
conversion since the second converter is unipolar.
OPERATING
• The D/A converter and its reference are accurate to the
total number of bits desired for the final conversion (the A/D
converter need only be accurate to 6 bits).
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
VDD - VSS to exceed the absolute maximum rating.
Operating Voltage
The first converter can be offset-biased by adding a 20Ω
resistor at the bottom of the ladder and increasing the reference voltage by 1 LSB. If a 6.4V reference is used in the
system, for example, then the first CA3306 will require a
6.5V reference.
Input Signals
To prevent damage to the input protection circuit, input
signals should never be greater than VDD nor less than VSS .
Input currents must not exceed 20mA even when the power
supply is off. The zener (pin 4) is the only terminal allowed to
exceed VDD .
Definitions
Dynamic Performance Definitions
Fast Fourier Transform (FFT) techniques are used to evaluate
the dynamic performance of the converter. 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.
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.
Output Short Circuits
Shorting of outputs to VDD or VSS may damage CMOS
devices by exceeding the maximum device dissipation.
13
CA3306, CA3306A, CA3306C
Application Circuits
OF
B7
(MSB)
B6
V+
B6
B5
B5
OF
B4
B4
1K
RC
VSS
DATA
OUTPUT
0.1µF
CA3306
CLOCK
INPUT
VZ
B3
CE2
B2
B3
B2
(LSB)
CE1
B1
B1
V+
CLK
VSS
VDD
0.2µF
10µF
PH
VIN
VREF + VREF 0.1µF
ADJUST POT
TO 1/2 VZ
B6
B5
OF
B4
CA3306
RC
VSS
0.1µF
V+
VZ
CE2
B3
CE1
B2
CLK
B1
VDD
0.2µF
PH
10µF
NOTE:
VDD MUST BE ≥ VZ FOR CIRCUIT TO WORK
WITH VZ CONNECTED TO VREF+
VIN
VREF -
SIGNAL
INPUT
VREF +
0.1µF
FIGURE 17. TYPICAL CA3306 7-BIT RESOLUTION CONFIGURATION
14
CA3306, CA3306A, CA3306C
Application Circuits
(Continued)
(MSB)
B6
V+
B6
B5
B5
OF
B4
B4
DATA
OUTPUT
RC
VSS
0.1µF
CA3306
CLOCK
INPUT
VZ
B3
CE2
B2
B3
B2
(LSB)
CE1
B1
B1
V+
CLK
VSS
VDD
0.2µF
10µF
PH
VIN
VREF + VREF 0.1µF
ADJUST POT
TO 1/2 VZ
B6
B5
OF
B4
CA3306
VSS
RC
0.1µF
V+
VZ
B3
CE2
B2
CE1
CLK
B1
VDD
0.2µF
V+
PH
10µF
NOTE:
VDD MUST BE ≥ VZ FOR CIRCUIT TO WORK
WITH VZ CONNECTED TO VREF+
VIN
VREF -
SIGNAL
INPUT
VREF +
0.1µF
FIGURE 18. TYPICAL CA3306 6-BIT RESOLUTION CONFIGURATION WITH DOUBLE SAMPLING RATE CAPABILITY
15
Application Circuits
(Continued)
BINARY
ADDER
B12
B6’
NO. 1
6-BIT
FLASH
ADC
S/H, VIN
B1’
B6 + 0
B5 + 0
B4 + 0
B3 + 0
B2 + 0
B1 + B7
B6
+
∑
-
6-BIT DAC
(12 BIT ACCURACY)
X32
B6
NO. 2
6-BIT
FLASH
ADC
B1
B1
B7
NO. 3
6-BIT
FLASH
ADC
B6
B1
CONTROL
LOGIC
FIGURE 19. TYPICAL CA3306, 800ns, 12-BIT ADC SYSTEM
16
Application Circuits
5V
(Continued)
VDD
CA3306
VDD
CA3306 INPUT
TYPICAL FOR:
1K
CA3306
CLK
50kΩ
5V
CA3306 INPUTS
TYPICAL FOR:
0.01µF
PHASE
CE1
CE2
74LS04
7406
OPEN COLLECTOR DRIVER
CA3306 OUTPUTS
TYPICAL FOR:
CA3306
VDD
5V
B1
B2
B3
B4
B5
B6
OF
CD74HC 4049 (INV.), OR
CD74HC4050 (NON-INV.), OR
ANY LOW POWER SCHOTTKY
TTL WITH HIGH INPUT VOLTAGE
RATING (MANY LS DEVICES ARE
RATED TO ACCEPT VOLTAGES
UP TO 15V).
FIGURE 20. 5V LOGIC INTERFACE CIRCUIT FOR VDD > 5.5V
All Intersil U.S. products are manufactured, assembled and tested utilizing ISO9000 quality systems.
Intersil Corporation’s quality certifications can be viewed at www.intersil.com/design/quality
Intersil products are sold by description only. Intersil Corporation reserves the right to make changes in circuit design, software 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 www.intersil.com
17