INTERSIL HI5800_00

HI5800
®
Data Sheet
July 2000
File Number
2938.2
12-Bit, 3MSPS, Sampling A/D Converter
Features
The HI5800 is a monolithic, 12-bit, sampling Analog-toDigital Converter fabricated in the HBC10 BiCMOS process.
It is a complete subsystem containing a sample and hold
amplifier, voltage reference, two-step subranging A/D, error
correction, control logic, and timing generator. The HI5800 is
designed for high speed applications where wide bandwidth,
accuracy and low distortion are essential.
• Throughput Rate . . . . . . . . . . . . . . . . . . . . . . . . . . 3MSPS
• Technical Brief TB363 “Guidelines for Handling and
Processing Moisture Sensitive Surface Mount Devices
(SMDs)”
HI5800BID
±1 LSB
HI5800JCD
HI5800KCD
±2 LSB
±1 LSB
HI5800-EV
• Buffered Sample and Hold Amplifier
• Precision Voltage Reference
• 20MHz Input BW Allows Sampling Beyond Nyquist
• Zero Latency/No Pipeline Delay
Applications
• High Speed Data Acquisition Systems
Ordering Information
LINEARITY
• Integral Linearity Error . . . . . . . . . . . . . . . . . . . . . 1.0 LSB
• Input Signal Range. . . . . . . . . . . . . . . . . . . . . . . . . . ±2.5V
Related Literature
PART
NUMBER
• 12-Bit, No Missing Codes Over Temperature
TEMP.
RANGE
(oC)
• Medical Imaging
PACKAGE
-40 to 85 40 Ld SBDIP
0 to 70
PKG.
NO.
40 Ld SBDIP
D40.6
D40.6
• Radar Signal Analysis
• Document and Film Scanners
• Vibration/Waveform Spectrum Analysis
• Digital Servo Control
25
Evaluation Board
Pinout
HI5800
(SBDIP)
TOP VIEW
REFIN 1
40 IRQ
ROADJ 2
39 OVF
RGADJ 3
38 AVCC
AVCC 4
REFOUT 5
36 D10
VIN 6
35 D9
AGND 7
34 D8
ADJ+ 8
33 DVCC
ADJ- 9
32 DGND
AVEE 10
31 AGND
AVCC 11
30 AVEE
AGND 12
29 D7
AVEE 13
28 D6
A0 14
27 D5
CS 15
26 D4
OE 16
25 D3
CONV 17
DVEE 18
24 D2
DGND 19
22 D0 (LSB)
21 AVCC
DVCC 20
4-1
37 D11 (MSB)
23 D1
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
HI5800
Functional Block Diagram
REFIN
REFERENCE
7-BIT
LATCH
D1
OUTPUT LATCHES
7-BIT
LATCH
REFOUT
ERROR CORRECTION
D0 (LSB)
D2
DIGITAL
OUTPUTS
D10
D11 (MSB)
OVF
ADJ+
ADJ-
IRQ
VIN
7-BIT
DAC
7-BIT
FLASH
S AND H
CS
CONTROL
LOGIC
AND
TIMING
CONV
OE
AO
X32
AVCC
AVEE
DVCC
DVEE
AGND
RGADJ
DGND
ROADJ
Typical Application Schematic
C23
+10µF
C22
0.1µF
C1
0.01µF
HI5800
(22) (LSB) D0
D0
(23) D1
D1
REF_IN (1)
(24) D2
D2
REF_OUT (5)
(25) D3
D3
(26) D4
D4
(27) D5
D5
(28) D6
D6
(29) D7
D7
(34) D8
D8
(35) D9
(36) D10
D10
AGND (7)
AGND (12)
AGND (31)
DGND (19)
DGND (32)
VIN
VIN (6)
CONV
D11
(40) IRQ
(39) OVF
IRQ
OVF
A0
CS
OE (16)
A0 (14)
CS (15)
GND
10µF, 0.1µF, AND 0.01µF CAPS ARE PLACED
AS CLOSE TO PART AS POSSIBLE
D9
(37) (MSB) D11
CONV (17)
OE
BNC
(18) DVEE
(33) DVCC
(20)
0.1µF
+
10µF
0.1µF
(4) AVCC
(11)
AVCC
R9
10K
RO_ADJ (2)
R10
10K
RG_ADJ (3)
(21) AVCC
(38)
ADJ+ (8)
(10) AVEE
(13) (30)
AVEE
R11
10K
ADJ- (9)
4-2
0.1µF
0.1µF
10µF
+
+
10µF
+
10µF
HI5800
Absolute Maximum Ratings
Thermal Information
Supply Voltages
AVCC or DVCC to GND . . . . . . . . . . . . . . . . . . . . . . . . . . . . +5.5V
AVEE or DVEE to GND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . -5.5V
DGND to AGND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ±0.3V
Analog Input Pins
Reference Input REFIN . . . . . . . . . . . . . . . . . . . . . . . . . . . +2.75V
Signal Input V IN . . . . . . . . . . . . . . . . . . . . . . . . . . ±(REFIN +0.2V)
ROADJ , RGADJ , ADJ+, ADJ- . . . . . . . . . . . . . . . . . . . . VEE to VCC
Digital I/O Pins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GND to VCC
Thermal Resistance (Typical, Note 1)
θJA ( oC/W)
θJC (oC/W)
SBDIP Package . . . . . . . . . . . . . . . . . .
40
15
Maximum Junction Temperature
SBDIP Package . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175oC
Maximum Storage Temperature Range . . . . . . . . . -65oC to 150oC
Maximum Lead Temperature (Soldering, 10s) . . . . . . . . . . . . 300oC
Operating Conditions
Temperature Range
HI5800JCD/KCD . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0oC to 70oC
HI5800BID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -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
PARAMETER
AVCC = +5V, DVCC = +5V, AVEE = -5V, DVEE = -5V; Internal Reference Used,
Unless Otherwise Specified
TEST CONDITIONS
HI5800JCD
HI5800KCD, HI5800BID
0oC TO 70oC
0oC TO 70oC
-40oC TO 85oC
MIN
TYP
MAX
MIN
TYP
MAX
UNITS
12
-
-
12
-
-
Bits
SYSTEM PERFORMANCE
Resolution
Integral Linearity Error, INL
fS = 3MHz, fIN = 45Hz Ramp
-
±0.7
±2
-
±0.5
±1
LSB
Differential Linearity Error, DNL
(Guaranteed No Missing Codes)
fS = 3MHz, fIN = 45Hz Ramp
-
±0.5
±1
-
±0.3
±1
LSB
Offset Error, VOS
(Adjustable to Zero)
(Note 8)
JCD, KCD
-
±2
±15
-
±2
±15
LSB
BID
-
-
-
-
±3
±15
LSB
Full Scale Error, FSE
(Adjustable to Zero)
(Note 8)
JCD, KCD
-
±2
±15
-
±2
±15
LSB
BID
-
-
-
-
±3
±15
LSB
DYNAMIC CHARACTERISTICS (Input Signal Level 0.5dB Below Full Scale)
Throughput Rate
No Missing Codes
3.0
-
-
3.0
-
-
MSPS
Signal to Noise Ratio (SNR)
fS = 3MHz, fIN = 20kHz
66
69
-
68
71
-
dB
RMS Signal
= ------------------------------RMS Noise
fS = 3MHz, fIN = 1MHz
65
67
-
67
69
-
dB
Signal to Noise Ratio (SINAD)
fS = 3MHz, fIN = 20kHz
66
68
-
68
71
-
dB
RMS Signal
= -------------------------------------------------------------RMS Noise + Distortion
fS = 3MHz, fIN = 1MHz
65
67
-
67
68
-
dB
Total Harmonic Distortion, THD
fS = 3MHz, fIN = 20kHz
-
-74
-70
-
-85
-74
dBc
fS = 3MHz, fIN = 1MHz
-
-70
-68
-
-77
-70
dBc
Spurious Free Dynamic Range,
SFDR
fS = 3MHz, fIN = 20kHz
71
76
-
76
86
-
dBc
fS = 3MHz, fIN = 1MHz
68
72
-
71
77
-
dBc
Intermodulation Distortion, IMD
fS = 3MHz, f1 = 49kHz,
f2 = 50kHz (Note 3)
-
-74
-66
-
-79
-70
dBc
Differential Gain
fS = 1MHz
-
0.9
-
-
0.9
-
%
Differential Phase
fS = 1MHz
-
0.05
-
-
0.05
-
Degrees
Aperture Delay, tAD
(Note 3)
-
12
20
-
12
20
ns
4-3
HI5800
Electrical Specifications
AVCC = +5V, DVCC = +5V, AVEE = -5V, DVEE = -5V; Internal Reference Used,
Unless Otherwise Specified (Continued)
PARAMETER
TEST CONDITIONS
HI5800JCD
HI5800KCD, HI5800BID
0oC TO 70oC
0oC TO 70oC
-40oC TO 85oC
MIN
TYP
MAX
MIN
TYP
MAX
UNITS
-
10
20
-
10
20
ps
Input Voltage Range
-
±2.5
±2.7
-
±2.5
±2.7
V
Input Resistance
1
3
-
1
3
-
MΩ
Input Capacitance
-
5
-
-
5
-
pF
Input Current
-
±1
±10
-
±1
±10
µA
Input Bandwidth
-
20
-
-
20
-
MHz
2.450
2.500
2.550
2.470
2.500
2.530
V
2
-
-
2
-
-
mA
-
20
-
-
13
-
ppm / oC
Aperture Jitter, tAJ
(Note 3)
ANALOG INPUT
INTERNAL VOLTAGE REFERENCE
Reference Output Voltage,
REFOUT (Loaded)
Reference Output Current
(Note 5)
Reference Temperature
Coefficient
REFERENCE INPUT
Reference Input Range
-
2.5
2.6
-
2.5
2.6
V
Reference Input Resistance
-
200
-
-
200
-
Ω
2.0
-
-
2.0
-
-
V
DIGITAL INPUTS
(Note 6)
Input Logic High Voltage, VIH
-
-
0.8
-
-
0.8
V
Input Logic Current, IIL
VIN = 0V, 5V
-
±±1
±10
-
±1
±10
µA
Digital Input Capacitance, CIN
VIN = 0V
-
5
-
-
5
-
pF
Input Logic Low Voltage, V IL
DIGITAL OUTPUTS
Output Logic High Voltage, VOH
IOUT = -160µA
2.4
4.3
-
2.4
4.3
-
V
Output Logic Low Voltage, V OL
IOUT = 3.2mA
-
0.22
0.4
-
0.22
0.4
V
Output Logic High Current, IOH
-0.160
-6
-
-0.160
-6
-
mA
Output Logic Low Current, IOL
3.2
6
-
3.2
6
-
mA
-
±±1
±10
-
±1
±10
µA
-
10
-
-
10
-
pF
VOUT = 0V, 5V
Output Three-State Leakage
Current, IOZ
Digital Output Capacitance, COUT
TIMING CHARACTERISTICS
Minimum CONV Pulse, t1
(Notes 3, 4)
10
-
-
10
-
-
ns
CS to CONV Setup Time, t2
(Note 3)
10
-
-
10
-
-
ns
CONV to CS Setup Time, t3
(Note 3)
0
-
-
0
-
-
ns
Minimum OE Pulse, t4
(Notes 3, 5)
15
-
-
15
-
-
ns
CS to OE Setup Time, t5
(Note 3)
0
-
-
0
-
-
ns
OE to CS Setup Time, t6
(Note 3)
0
-
-
0
-
-
ns
IRQ Delay from Start Convert, t7
(Note 3)
JCD, KCD
IRQ Pulse Width, t8
BID
Minimum Cycle Time for
Conversion, t9
10
20
25
10
20
25
ns
190
200
230
190
200
230
ns
-
-
-
180
195
230
ns
-
325
333
-
325
333
ns
IRQ to Data Valid Delay, t10
(Note 3)
-5
0
+5
-5
0
+5
ns
Minimum A0 Pulse, t11
(Notes 3, 5)
10
-
-
10
-
-
ns
4-4
HI5800
Electrical Specifications
AVCC = +5V, DVCC = +5V, AVEE = -5V, DVEE = -5V; Internal Reference Used,
Unless Otherwise Specified (Continued)
PARAMETER
TEST CONDITIONS
HI5800JCD
HI5800KCD, HI5800BID
0oC TO 70oC
0oC TO 70oC
-40oC TO 85oC
MIN
TYP
MAX
MIN
TYP
MAX
UNITS
Data Access from OE Low, t12
(Note 3)
10
18
25
10
18
25
ns
LSB, Nibble Delay from A0 High, t13
(Note 3)
-
10
20
-
10
20
ns
MSB Delay from A0 Low, t14
(Note 3)
-
14
20
-
14
20
ns
CS to Float Delay, t15
(Note 3)
10
18
25
10
18
25
ns
Minimum CS Pulse, t16
(Notes 3, 5)
15
-
-
15
-
-
ns
CS to Data Valid Delay, t17
(Note 3)
10
18
25
10
18
25
ns
Output Fall 2 Time, tf
(Note 3)
-
5
20
-
5
20
ns
Output Rise Time, tr
(Note 3)
-
5
20
-
5
20
ns
IVCC
-
170
220
-
170
220
mA
IVEE
-
150
190
-
150
190
mA
IDVCC
-
24
40
-
24
40
mA
POWER SUPPLY CHARACTERISTICS
IDVEE
-
2
5
-
2
5
mA
Power Dissipation
-
1.7
2.2
-
1.7
2.2
W
-
0.01
-
-
±0.01
-
% /%
V CC, VEE ±5%
PSRR
NOTES:
2. Dissipation rating assumes device is mounted with all leads soldered to printed circuit board.
3. Parameter guaranteed by design or characterization and not production tested.
4. Recommended pulse width for CONV is 60ns.
5. Recommended minimum pulse width is 25ns.
6. This is the additional current available from the REFOUT pin with the REFOUT pin driving the REFIN pin.
7. The A0 pin VIH at -40oC may exceed 2.0V by up to 0.4V at initial power up.
8. Excludes error due to internal reference temperature drift.
Timing Diagrams
CONV
t1
CS
t16
t7
IRQ
ACQUIRE N
N CONVERSION
t8
N DATA
N - 1 DATA
DATA VALID
AO
OE
t15
D0 - D11, OVF
N DATA
t12
FIGURE 1. SINGLE SHOT TIMING
4-5
t17
HI5800
Timing Diagrams
(Continued)
CS
CS
t2
t5
t3
CONV
t6
OE
t1
t4
FIGURE 2A. START CONVERSION SETUP TIME
FIGURE 2B. OUTPUT ENABLE SETUP TIME
CONV
CS
t7
IRQ
N CONVERSION
t8
ACQUIRE N
N+1 CONVERSION
ACQUIRE N + 1
t10
t9
DATA VALID
N DATA
N - 1 DATA
N + 1 DATA
t11
AO
OE
t13
D4 - D11
t14
D11 - D4
D3 - D0, 0000
D0 - D11, OVF
D11 - D4
D11 - D4
N + 1 DATA
N DATA
t12
FIGURE 3. CONTINUOUS CONVERSION TIMING
Typical Performance Curves
80
90
70
80
70
60
60
dB
dB
50
40
30
50
40
30
20
20
10
10
0
20K
200K
INPUT FREQUENCY (Hz)
FIGURE 4. TYPICAL SNR vs INPUT FREQUENCY
4-6
2M
0
20K
200K
INPUT FREQUENCY (Hz)
FIGURE 5. TYPICAL THD vs INPUT FREQUENCY
2M
HI5800
Typical Performance Curves
(Continued)
80
90
70
80
60
70
60
dB
dB
50
40
30
50
40
30
20
20
10
10
0
20K
200K
0
20K
2M
200K
FIGURE 6. TYPICAL SINAD vs INPUT FREQUENCY
FIGURE 7. TYPICAL SFDR vs INPUT FREQUENCY
11.5
12
11.25
10
11.0
ENOB
dB
8
6
4
10.75
10.5
10.25
2
10.0
9.75
0
20K
200K
0.5
2M
1.0
INPUT FREQUENCY (Hz)
1.50
VREF (V)
2.00
2.50
FIGURE 8. TYPICAL EFFECTIVE NUMBER OF BITS vs INPUT
FREQUENCY
FIGURE 9. EFFECTIVE NUMBER OF BITS vs REFERENCE
VOLTAGE (f S = 3MHz, fIN = 20kHz)
1.0
1.0
0.5
0.5
INL ERROR (LSB)
DNL ERROR (LSB)
2M
INPUT FREQUENCY (Hz)
INPUT FREQUENCY (Hz)
0
0
-0.5
-0.5
-1.0
-1.0
0
1000
2000
3000
CODE
FIGURE 10. DIFFERENTIAL NON-LINEARITY
4-7
4000
0
1000
2000
3000
CODE
FIGURE 11. INTEGRAL NON-LINEARITY
4000
HI5800
(Continued)
10
0
10
-25
-25
0
OUTPUT LEVEL (dB)
OUTPUT LEVEL (dB)
Typical Performance Curves
-50
-75
-100
365K
730K
1.095M
-75
-100
-125
-135
-125
-135
0
-50
1.46M
0
365K
730K
10
10
0
0
-25
-25
-50
-75
-100
0
365K
730K
1.095M
-50
49K
50K
-75
-100
-125
-135
1.46M
0
18.3K
36.6K
55K
73.3K
91.6K
110K
FREQUENCY (Hz)
FREQUENCY (Hz)
FIGURE 14. FFT SPECTRAL PLOT FOR fIN = 2MHz, f S = 3MHz
FIGURE 15. INTERMODULATION DISTORTION PLOT FOR
fIN = 49kHz, 50kHz at fS = 3MHz
Pin Descriptions
PIN #
SYMBOL
PIN DESCRIPTION
1
REFIN
External Reference Input.
2
ROADJ
DAC Offset Adjust (Connect to AGND If Not Used).
3
RGADJ
DAC Gain Adjust (Connect to AGND If Not Used).
4
AV CC
Analog Positive Power Supply, +5V.
5
REFOUT
-
NC
No Connection.
6
VIN
Analog Input Voltage.
7
AGND
Analog Ground.
8
ADJ+
Sample/Hold Offset Adjust (Connect to AGND If Not Used).
9
ADJ-
Sample/Hold Offset Adjust (Connect to AGND If Not Used).
Internal Reference Output, +2.5V.
4-8
1.46M
FIGURE 13. FFT SPECTRAL PLOT FOR fIN = 1MHz, fS = 3MHz
OUTPUT LEVEL (dB)
OUTPUT LEVEL (dB)
FIGURE 12. FFT SPECTRAL PLOT FOR fIN = 20kHz, fS = 3MHz
-125
-135
1.095M
FREQUENCY (Hz)
FREQUENCY (Hz)
HI5800
Pin Descriptions
(Continued)
PIN #
SYMBOL
PIN DESCRIPTION
10
AVEE
Analog Negative Power Supply, -5V.
11
AV CC
Analog Positive Power Supply, +5V.
12
AGND
Analog Ground.
13
AVEE
Analog Negative Power Supply, -5V.
14
A0
Output Byte Control Input, active low. When low, data is presented as a 12-bit word or the upper byte (D11 - D4) in 8-bit
mode. When high, the second byte contains the lower LSBs (D3 - D0) with 4 trailing zeroes. See Text.
15
CS
Chip Select Input, active low. Dominates all control inputs.
-
NC
No Connection.
16
OE
Output Enable Input, active low.
17
CONV
Convert Start Input. Initiates conversion on the falling edge. If held low, continuous conversion mode overrides and
remains in effect until the input goes high.
18
DVEE
Digital Negative Power Supply, -5V.
19
DGND
Digital Ground.
20
DVCC
Digital Positive Power Supply, +5V.
21
AV CC
Analog Positive Power Supply, +5V.
22
D0
Data Bit 0, (LSB).
23
D1
Data Bit 1.
24
D2
Data Bit 2.
25
D3
Data Bit 3.
-
NC
No Connection
26
D4
Data Bit 4.
27
D5
Data Bit 5.
28
D6
Data Bit 6.
29
D7
Data Bit 7.
30
AVEE
Analog Negative Power Supply, -5V.
31
AGND
Analog Ground.
32
DGND
Digital Ground.
33
DVCC
Digital Positive Power Supply, +5V.
34
D8
Data Bit 8.
35
D9
Data Bit 9.
-
NC
No Connection.
36
D10
Data Bit 10.
37
D11
Data Bit 11 (MSB).
38
AV CC
39
OVF
Overflow Output. Active high when either an overrange or underrange analog input condition is detected.
40
IRQ
Interrupt ReQuest Output. Goes low when a conversion is complete.
Analog Positive Power Supply, +5V.
4-9
HI5800
Description
The HI5800 is a 12-bit, two-step, sampling analog-to-digital
converter which uses a subranging technique with digital
error correction. As illustrated in the block diagram, it uses a
sample and hold front end, 7-bit, R-2R D/A converter which
is laser trimmed to 14 bits accuracy, a 7-bit BiCMOS flash
converter, precision bandgap reference, digital controller
and timing generator, error correction logic, output latches
and BiCMOS output drivers.
The falling edge of the convert command signal puts the
sample and hold (S/H) in the hold mode and the conversion
process begins. At this point the Interrupt Request (IRQ) line is
set high indicating that a conversion is in progress. The output
of the S/H circuit drives the input of the 7-bit flash converter
through a switch. After allowing the flash to settle, the
intermediate output of the flash is stored in the latches which
feed the D/A and error correction logic. The D/A reconstructs
the analog signal and feeds the gain amplifier whose summing
node subtracts the held signal of the S/H and amplifies the
residue by 32. This signal is then switched to the flash for a
second pass using the input switch. The output of the second
flash conversion is fed directly to the error correction which
reconstructs the twelve bit word from the fourteen bit input. The
logic also decodes the overflow bit and the polarity of the
overflow. The output of the error correction is then gated
through the read controller to the output drivers. The data is
ready on the bus as soon as the IRQ line goes low.
I/O Control Inputs
The converter has four active low inputs (CS, CONV, OE
and A0) and fourteen outputs (D0 - D11, IRQ and OVF). All
inputs and outputs are TTL compatible and will also interface
to the newer TTL compatible families. All four inputs are
CMOS high input impedance stages and all outputs are
BiMOS drivers capable of driving 100pF loads.
In order to initiate a conversion or read the data bus, CS should
be held low. The conversion is initiated by the falling edge of the
CONV command. The OE input controls the output bus directly
and is independent of the conversion process. The data on the
bus changes just before the IRQ goes low. Therefore if the OE
line is held low all the time, the data on the bus will change just
before the IRQ line goes low. The byte control signal A0 is also
independent of the conversion process and the byte can be
manipulated anytime. When A0 is low the 12-bits and overflow
word is read on the bus. The bus can also be hooked up such
that the upper byte (D11 - D4) is read when A0 is low. When A0
is high, the lower byte (D3 - D0) is output on the same eight
pins with trailing zeros.
In order to minimize switching noise during a conversion,
byte manipulations done using the A0 signal should be done
in the single shot mode and A0 should be changed during
the acquisition phase. For accuracy, allow sufficient time for
settling from any glitches before the next conversion.
Once a conversion is started, the converter will complete the
conversion and acquisition periods irrespective of the input
4-10
states. If during these cycles another convert command is
issued, it will be ignored until the acquire phase is complete.
Stand Alone Operation
The converter can be operated in a stand alone configuration
with bus inputs controlling the converter. The conversion will be
started on the negative edge of the convert (CONV) pulse as
long as this pulse is less than the converter throughput rate. If
the converter is given multiple convert commands, it will ignore
all but the first command until such time when the acquisition
period of the next cycle is complete. At this point it will start a
new conversion on the first negative edge of the input
command. This allows the converter to be synchronized to a
multiple of a faster external clock. The new output data of the
conversion is available on the same cycle at the negative edge
of the IRQ pulse and is valid until the next negative edge of the
IRQ pulse. Data may be accessed at any time during these
cycles. It should be noted that if the data bus is kept enabled all
the time (OE is low), then the data will be updating just before
the IRQ goes low. During this time, the data may not be valid
for a few nanoseconds.
Continuous Convert Mode
The converter can be operated at its maximum rate by taking
the CONV line low (supplying the first negative edge) and
holding it low. This enables the continuous convert mode.
During this time, at the end of the internal acquisition period,
the converter automatically starts a new conversion. The
data will be valid between the IRQ negative edges.
Note that there is no pipeline delay on the data. The output data
is available during the same cycle as the conversion and is
valid until the next conversion ends. This allows data access to
both previous and present conversions in the same cycle.
When initiating a conversion or a series of conversions, the
last signal (CS and CONV) to arrive dominates the function.
The same condition holds true for enabling the bus to read
the data (CS and OE). To terminate the bus operations, the
first signal (CS and OE) to arrive dominates the function.
Interrupt Request Output
The interrupt request line (IRQ) goes high at the start of each
conversion and goes low to indicate the start of the acquisition.
During the time that IRQ is high, the internal sample and hold is
in hold mode. At the termination of IRQ, the sample and hold
switches to acquire mode which lasts approximately 100ns. If
no convert command is issued for a period of time, the sample
and hold simply remains in acquire mode tracking the analog
input signal until the next conversion cycle is initiated. The IRQ
line is the only output that is not three-stateable.
Analog Input, VIN
The analog input of the HI5800 is coupled into the input
stage of the Sample and Hold amplifier. The input is a high
impedance bipolar differential pair complete with an ESD
protection circuit. Typically it has >3MΩ input impedance.
With this high input impedance circuit, the HI5800 is easily
HI5800
interfaced to any type of op amp without a requirement for a
high drive capability. Adequate precautions should be taken
while driving the input from high voltage output op amps to
ensure that the analog input pin is not overdriven above the
specified maximum limits. For a +2.5V reference, the analog
input range is ±2.5V. This input range scales with the value
of the external reference voltage if the internal reference is
not used. For best performance, the analog ground pin next
to the analog input should be utilized for signal return.
Figures 16 and 17 illustrate the use of an input buffer as a
level shifter to convert a unipolar signal to the bipolar input
used by the HI5800. Figure 16 is an example of a noninverting buffer that takes a 0 to 2.5V input and shifts it to
±2.5V. The gain can be calculated from:
V
OUT
is recommended that the output of the reference be decoupled
with good quality capacitors to reduce the high frequency noise.
Reference Input, REFIN
The converter requires a voltage reference connected to the
REFIN pin. This can be the above internal reference or it can
be an external reference. It is recommended that adequate
high frequency decoupling is provided at the reference input
pin in order to minimize overall converter noise.
A user trying to provide an external reference to a HI5800 is
faced with two problems. First, the drift of the reference over
temperature must be very low. Second, it must be capable of
driving the 200Ω input impedance seen at the REFIN pin of
the HI5800. Figure 18 is a recommended circuit for doing
this that is capable of 2ppm/ oC drift over temperature.
R1
R2
= 1 + ------------------------- × V – ---------------------- × V
OFFSET
IN R1 + R3
( R1|| R3 )
HA5177 HA5002
+15
+15
VOUT VFB
R1|| R
R1R3
= ---------------------3
R1 + R3
+15V
+
10µ
R2
1kΩ
VOFFSET
VOUT
+
VIN
HI5800
VIN
HA2841
-15V
0.1
FIGURE 16. NON-INVERTING BUFFER
Figure 17 is an example of an inverting buffer that level shifts
a 0V to 5V input to ±2.5V. Its gain can be calculated from:
V
VIN
OUT
= ( – R2 ⁄ R1 ) × V
– ( R2 ⁄ R3 ) × V
R1
1kΩ
R3
VOFFSET
IN
OFFSET
.
R2
+15V
-
2kΩ
+
1kΩ
0.1
HA2841
VOUT
HI5800
VIN
-15V
0.1
FIGURE 17. INVERTING BUFFER
Note that the correct op amp must be chosen in order to not
degrade the overall dynamic performance of the circuit.
Recommended op amps are called out in the figures.
Voltage Reference, REFOUT
The HI5800 has a curvature corrected internal band-gap
reference generator with a buffer amplifier capable of driving up
to 15mA. The band-gap and amplifier are trimmed to give
+2.50V. When connected to the reference input pin REFIN , the
reference is capable of driving up to 2mA externally. Further
loading may degrade the performance of the output voltage. It
4-11
C
RB
10
kΩ
-
10µ
-15
+
0.1
-15
FIGURE 18. EXTERNAL REFERENCE
0.1
R1
2kΩ
0.1
RA
HI5800
REFIN
10Ω
+
LOW TC RESISTOR
+15V
R3
2kΩ
REF101
Supply and Ground Considerations
The HI5800 has separate analog and digital supply and
ground pins to help keep digital noise out of the analog signal
path. For the best performance, the part should be mounted
on a board that provides separate low impedance planes for
the analog and digital supplies and grounds. Only connect the
two grounds together at one place preferably as close as
possible to the part. The supplies should be driven by clean
linear regulated supplies. The board should also have good
high frequency decoupling capacitors mounted as close as
possible to the HI5800.
If the part is powered off a single supply then the analog
supply and ground pins should be isolated by ferrite beads
from the digital supply and ground pins.
Also, it is recommended that the turn-on power supply
sequencing be such that the analog positive supply, AICC ,
come up first, followed by the remaining supplies.
Refer to the Application Note “Using Intersil High Speed A/D
Converters” (AN9214) for additional suggestions to consider
when using the HI5800.
Error Adjustments
For most applications the accuracy of the HI5800 is sufficient
without any adjustments. In applications where accuracy is of
utmost importance three external adjustments are possible:
S/H offset, D/A offset and D/A gain. Figure 19 illustrates the use
of external potentiometers to reduce the HI5800 errors to zero.
The D/A offset (ROADJ) and S/H offset (ADJ+ and ADJ-)
trims adjust the voltage offset of the transfer curve while the
HI5800
D/A gain trim (RGADJ) adjusts the tilt of the transfer curve
around the curve midpoint (code 2048). The 10kΩ
potentiometers can be installed to achieve the desired
adjustment in the following manner.
Typically only one of the offset trimpots needs to be used.
The offset should first be adjusted to get code 2048 centered
at a desired DC input voltage such as 0V. Next the gain trim
can be adjusted by trimming the gain pot until the 4094 to
4095 code transition occurs at the desired voltage
(2.500V - 1.5 LSBs for a 2.5V reference). The gain trim can
also be done by adjusting the gain pot until the code 0 to 1
transition occurs at a particular voltage (-2.5V + 0.5 LSBs for
a 2.5V reference). If a nonzero offset is needed, then the
offset pot can be adjusted after the gain trim is finished. The
gain trim is simplified if an offset trim to zero is done first with
a nonzero offset trim done after the gain trim is finished. The
D/A offset and S/H offset trimpots have an identical effect on
the converter except that the S/H offset is a finer resolution
trim. The D/A offset and D/A gain typically have an
adjustment range of ±30 LSBs and the S/H offset typically
has an adjustment range of ± 20 LSBs.
VCC
ROADJ
10kΩ
10kΩ
RGADJ
VEE
ADJ+
10kΩ
VEE
ADJ-
FIGURE 19. D/A OFFSET, D/A GAIN AND S/H OFFSET
ADJUSTMENTS
TABLE 1. I/O TRUTH TABLE
INPUTS
OUTPUT
CS
CONV
OE
A0
IRQ
1
X
X
X
X
FUNCTION
0
0
X
X
X
Continuous convert mode.
0
X
0
0
X
Outputs all 12-bits and OVF or upper byte D11 - D4 in 8 bit mode.
0
X
0
1
X
In 8-bit mode, outputs lower LSBs D3 - D0 followed by 4 trailing zeroes
and OVF (See text).
0
1
X
X
0
Converter is in acquisition mode.
0
X
X
X
1
Converter is busy doing a conversion.
0
X
1
X
X
Data outputs and OVF in high impedance state.
No operation.
X’s = Don’t Care
TABLE 2. A/D OUTPUT CODE TABLE
CODE
DESCRIPTION
OUTPUT DATA (OFFSET BINARY)
LSB = 2 (REFIN)
4096
(NOTE)
INPUT
VOLTAGE
REFIN = 2.5V
(V)
OVF
D11
D10
D9
D8
D7
D6
D5
D4
D3
D2
D1
D0
≥+FS (Full Scale)
≥ +2.5000
1
1
1
1
1
1
1
1
1
1
1
1
1
+FS - 1 LSB
+2.49878
0
1
1
1
1
1
1
1
1
1
1
1
1
+3/4 FS
+1.8750
0
1
1
1
0
0
0
0
0
0
0
0
0
MSB
LSB
+1/2 FS
+1.2500
0
1
1
0
0
0
0
0
0
0
0
0
0
+1 LSB
+0.00122
0
1
0
0
0
0
0
0
0
0
0
0
1
0
0.0000
0
1
0
0
0
0
0
0
0
0
0
0
0
-1 LSB
-0.00122
0
0
1
1
1
1
1
1
1
1
1
1
1
-1/2 FS
-1.2500
0
0
1
0
0
0
0
0
0
0
0
0
0
-3/4 FS
-1.8750
0
0
0
1
0
0
0
0
0
0
0
0
0
-FS + 1 LSB
-2.49878
0
0
0
0
0
0
0
0
0
0
0
0
1
≤-FS
≤ -2.5000
1
0
0
0
0
0
0
0
0
0
0
0
0
NOTE: The voltages listed above represent the ideal center of each output code shown as a function of the reference voltage.
4-12
HI5800
If no external adjustments are required the following pins
should be connected to analog ground (AGND) for optimum
performance: ROADJ , RG ADJ , ADJ+, and ADJ-.
Typical Application Schematic
A typical application schematic diagram for the HI5800 is
shown with the block diagram. The adjust pins are shown
with 10kΩ potentiometers used for gain and offset
adjustments. These potentiometers may be left out and the
respective pins should be connected to ground for best
untrimmed performance.
Definitions
Static Performance Definitions
Offset, Full scale, and gain all use a measured value of the
internal voltage reference to determine the ideal plus and
minus full scale values. The results are all displayed in
LSBs.
Offset Error (VOS)
The first code transition should occur at a level 1/2 LSB
above the negative full scale. Offset is defined as the
deviation of the actual code transition from this point. Note
that this is adjustable to zero.
Full Scale Error (FSE)
The last code transition should occur for a analog input that
is 11/2 LSBs below positive full scale. Full scale error is
defined as the deviation of the actual code transition from
this point.
(decibels with respect to carrier) and DO NOT include any
correction factors for normalizing to full scale.
Signal-to-Noise Ratio (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.
Effective Number Of Bits (ENOB)
The effective number of bits (ENOB) is derived from the
SINAD data. ENOB is calculated from:
ENOB = (SINAD - 1.76 + V CORR) / 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.
Spurious Free Dynamic Range (SFDR)
SFDR is the ratio of the fundamental RMS amplitude to the
RMS amplitude of the next largest spur or spectral
component. If the harmonics are buried in the noise floor it is
the largest peak.
Intermodulation Distortion (IMD)
INL is the worst case deviation of a code center from a best
fit straight line calculated from the measured data.
Nonlinearities in the signal path will tend to generate
intermodulation products when two tones, f1 and f2 , are
present on the inputs. The ratio of the measured signal to
the distortion terms is calculated. The IMD products used to
calculate the total distortion are (f2-f1), (f2+f1), (2f1-f2),
(2f1+f2), (2f2-f1), (2f2+f1), (3f1-f2), (3f1+f2), (3f2-f1), (3f2+f1),
(2f2-2f1), (2f2+2f1), (2f1), (2f2), (2f1), (2f2), (4f1), (4f2). The
data reflects the sum of all the IMD products.
Power Supply Rejection (PSRR)
Full Power Input Bandwidth
Each of the power supplies are moved plus and minus 5%
and the shift in the offset and full scale error is noted. The
number reported is the percent change in these parameters
versus full scale divided by the percent change in the supply.
Full power input bandwidth is the frequency at which the
amplitude of the fundamental of the digital output word has
decreased 3dB below the amplitude of an input sine wave.
The input sine wave has a peak-to-peak amplitude equal to
the reference voltage. The bandwidth given is measured at
the specified sampling frequency.
Differential Linearity Error (DNL)
DNL is the worst case deviation of a code width from the
ideal value of 1 LSB. The converter is guaranteed for no
missing codes over all temperature ranges.
Integral Linearity Error (INL)
Dynamic Performance Definitions
Fast Fourier Transform (FFT) techniques are used to
evaluate the dynamic performance of the HI5800. 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. Distortion results are quoted in dBc
4-13
HI5800
Die Characteristics
DIE DIMENSIONS:
PASSIVATION:
202 mils x 283 mils x 19 mils
Type: Sandwich Passivation - Nitride +
Undoped Si Glass (USG)
Thickness: Nitride - 4KÅ, USG - 8KÅ, Total - 12kÅ ±2kÅ
METALLIZATION:
Metal 1: Type: AlSiCu, Thickness: 6kÅ +1500A/-750Å
Metal 2: Type: AlSiCu, Thickness: 16kÅ +2500A/1100Å
TRANSISTOR COUNT:
10K
SUBSTRATE POTENTIAL (POWERED UP):
VEE
Metallization Mask Layout
D10
D11 (MSB)
AVCC
OVF
IRQ
REF_IN
RO_ADJ
RG_ADJ
AVCC
AVCC
REF_OUT
REF_OUT
HI5800
VIN
AGND
D9
AGND
ADJ+
D8
ADJAVEE
AVEE
DVCC
AVCC
DGND
AVCC
AGND
VEE
AGND
AVEE
D7
AVEE
D6
A0
D5
CS
4-14
D3
D2
D1
D6 (LSB)
AVCC
DVCC
DVCC
DGND
DVEE
DGND
CONV
OE
D4
HI5800
Ceramic Dual-In-Line Metal Seal Packages (SBDIP)
c1
-A-
BASE
METAL
E
MIL-STD-1835 CDIP2-T40 (D-5, CONFIGURATION C)
40 LEAD CERAMIC DUAL-IN-LINE METAL SEAL PACKAGE
-BC A-B S
(c)
b1
M
(b)
M
bbb S
D40.6
LEAD FINISH
-D-
SECTION A-A
D S
D
BASE
PLANE
S2
SEATING
PLANE
Q
-C-
A
L
S1
eA
A A
b2
b
e
eA/2
c
aaa M C A - B S D S
ccc M C A - B S D S
NOTES:
1. Index area: A notch or a pin one identification mark shall be located adjacent to pin one and shall be located within the shaded
area shown. The manufacturer’s identification shall not be used
as a pin one identification mark.
2. The maximum limits of lead dimensions b and c or M shall be
measured at the centroid of the finished lead surfaces, when
solder dip or tin plate lead finish is applied.
3. Dimensions b1 and c1 apply to lead base metal only. Dimension
M applies to lead plating and finish thickness.
4. Corner leads (1, N, N/2, and N/2+1) may be configured with a
partial lead paddle. For this configuration dimension b3 replaces
dimension b2.
5. Dimension Q shall be measured from the seating plane to the
base plane.
6. Measure dimension S1 at all four corners.
7. Measure dimension S2 from the top of the ceramic body to the
nearest metallization or lead.
8. N is the maximum number of terminal positions.
9. Braze fillets shall be concave.
10. Dimensioning and tolerancing per ANSI Y14.5M - 1982.
11. Controlling dimension: INCH.
4-15
INCHES
SYMBOL
MIN
MILLIMETERS
MAX
MIN
MAX
NOTES
A
-
0.225
-
5.72
-
b
0.014
0.026
0.36
0.66
2
b1
0.014
0.023
0.36
0.58
3
b2
0.045
0.065
1.14
1.65
-
b3
0.023
0.045
0.58
1.14
4
c
0.008
0.018
0.20
0.46
2
c1
0.008
0.015
0.20
0.38
3
D
-
2.096
-
53.24
4
E
0.510
0.620
15.75
4
12.95
e
0.100 BSC
2.54 BSC
-
eA
0.600 BSC
15.24 BSC
-
eA/2
0.300 BSC
7.62 BSC
-
L
0.125
0.200
3.18
5.08
-
Q
0.015
0.070
0.38
1.78
5
S1
0.005
-
0.13
-
6
S2
0.005
-
0.13
-
7
α
90o
105o
90o
105o
-
aaa
-
0.015
-
0.38
-
bbb
-
0.030
-
0.76
-
ccc
-
0.010
-
0.25
-
M
-
0.0015
-
0.038
2
N
40
40
8
Rev. 0 4/94