N ANALOG
- W DEVICES
PrecisionSample-and-Hold
with 16-Channel
Multiplexer
III
.
FEA TURES
16 Single-Endedor 8 Differential
Mode Control
True 12-Bit Precision: Nonlinearity
Channels with Switchable
QO.005%
High Speed: 1OilS Acquisition Time to 0.01%
Complete and Calibrated: No Additional Parts Required
Small, Reliable: 32 Pin Hermetic Metal DIP
Versatile: Simple Interface to Popular Analog to Digital
Converters
High Differential Input Impedance (101°0) and Common
Mode Rejection (8OdB)
Fully Protected Multiplexer Inputs
OBS
~
OLE
PRODUCT DESCRIPTION
The AD362 is a complete, precision 16-channel data acquisition system analog input section in hybrid integrated circuit
form. Large-scale linear integrated circuitry, thick- and thinfilm technology and active laser trimming gives the AD362
extensive applications versatility along with full 12-bit
accuracy.
The AD362 contains two 9-channel multiplexers, a differential
amplifier, a sample-and-hold with high-speed output amplifier,
a channel address latch and control logic. The multiplexers may
be connected to the differential amplifier in either an 8-channel differential or 16-channel single-ended configuration. A
unique featUre of the AD362 is an internal user-controllable
analog switch that connects the multiplexers in either a singleended or differential mode. This allows a single device to perform in either mode without hard-wire programming and permits a mixtUre of single-ended and differential sources to be
interfaced by dynamically switching the input mode control.
The sample-and-hold mode control is designed to connect
direcdy to the "StatUs" output of an analog to digital converter so that a convert command to the ADC will automatically put the sample-and-hold into the "Hold" mode. A
precision hold capacitor is included with each AD362. The
AD362 output amplifier is capable of driving the unbuffered
analog input of most high-speed, 12-bit successive-approximation ADCs. Interface is thereby reduced to two simple connections with no additional components required.
When used with a 12-bit, 2S-microsecond ADC such as the
ADS72, ADS74 or AD ADC80, system throughput rate is as
high as 30kHz at full rated accuracy. The AD362KD is specified
for operation over a 0 to +70°C temperatUre range while the
~
Information furnished by Analog Devices is believed to be accurate
and reliable. However, no responsibility is assumed by Analog Devices
for its use; nor for any infringements of parents or other rights of third
parties which may result from its use. No license is granted by implication or otherwise under any patent or patent rights of Analog Devices.
---
AD362SD operates to specification from -55°C to + 125°C.
Processing to MIL-STD-883 , Class B is available for the
AD362SD. Both grades are packaged in a hermetic, electrostatically shielded 32-pin metal dual-in-line package.
PRODUCT HIGHLIGHTS
1. The AD362, when used with a precision analog to digital
converter, forms a complete, accurate, high-speed data
acquisition system.
2. The 16-input channels may be configured in single-ended,
differential or a mixtUre of both modes. Mode switching is
provided by a user-controllable internal analog switch.
3. Multiplexers, differential amplifier, sample-and-hold and
high-speed output buffer provide complete analog interfacing capabilities.
4. Internal channel address latches are provided to facilitate
interfacing the AD362 to data, address or control buses.
5. All grades of the AD362 are hermetically sealed in rugged
metal DIP packages.
6. A precision hold capacitor is provided with each AD362.
7. The AD362SD is specified over the entire military temperatUre range, -55°C to +12SoC. Processing to MIL-STD883, Class B is available.
TE
Route 1 Industrial Park; P.O. Box 280; Norwood, Mass. 02062
Tel: 617/329-4700
TWX: 710/394-6577
West Coast
Mid-West
Texas
714/842-1717
312/894-3300
214/231.5094
-
(typical@+25°C,:t15Vand+5V with 2000pF hold capacitorasprovided
SPECIFICATIONS
unlessotherwisenoted)
MODEL
AD362KD
AD362SD/AD362SD-883BI
16 Single-Ended or 8 Differential
(Electronically Selectable)
*
ANALOG INPUTS
Number of Inputs
Input Voltage Range, Linear
Tmin to Tmax
Input (Bias) Current, Per Channel
Input Impedance
On Channel.
Off Channel
Input Fault Current (Power Off or On)
Common Mode Rejection
Differential Mode
Mux Crosstalk (Interchannel,
Any Off Channel to Any On Channel)
Offset, Channel to Channel
*
*
:tIOV min
:tSOnA max
IOl0n, IOOpF
1010 n, IOpF
20mA, max, Internally Limited
70dB min (80dB typ) @ 1kHz, 20V pop
-80dB max (-90dB typ) @ 1kHz, 20V p-p
:t2.SmV mIX
OBS
ACCURACY
Gain Error, Tmin to Tmax
Offset Error, Tmin to Tmax
Linearity Error
:to.02% FSR, max
:t4mV
:to.005% max
:to.Ol% max
ImV pop, 0.1 to IMHz, mIX
2m V p-p, 0.1 to 1MHz, mIX
Tmin to Tmax
TEMPERATURE COEFFICIENTS
Gain, Tmill to Tmax
Offset, :tlOV Range, Tmin to Tmax
SAMPLE AND HOLD DYNAMICS
ApertUre Delay
Aperture Uncertainty
Acquisition Time, for 20V Step to
:to.01 % of Final Value
Feedthrough
Droop Rate
DIGIT AL INPUT SIGNALS2
Input Channel Select (Pins 28-31)
:t4ppm/C max
:t2ppm/ C max
lOOns max (SOns typ)
SOOpsmIX (lOOps typ)
18J.LSmax (lO~s typ)
-70dB max (-80dB typ) @ 1kHz
2mV/ms max (lmV/ms typ)
*
*
*
*
*
*
*
*
*
:t2ppm/ C max
:t1.Sppm/oC mIX
*
*
*
*
*
4-Bit Binary, Channel Address
lLS TTL Load
"I ": Latch Transparent
"0": Latched
8LS TTL Loads
*
*
*
*
*
Single Ended/Differential
Mode Select (Pin 1)
"0": Single-Ended Mode
"I ": Differential Mode
3TTL Loads
*
*
*
Sample and Hold Command (Pin 13)
"0": Sample Mode
"I ": Hold Mode
1TTL Load
*
*
*
Channel Select Latch (Pin 32)
POWER REQUIREMENTS
Supply Voltages/Currents
+ISV, :!:S% @ 30mA mix
-ISV, :!:S% @ 30mA mix
+SV, :tS% @ 40mA mix
Total Power Dissipation
TEMPERATURE RANGE
Specification
Storage
NOTES:
1
*
*
*
OLE
Tmin to Tmax
Noise Error
1.1 Watts mix
*
0 to +700C
-SSoC to +8S0C3
-SSoC to +12SoC
-SSoC to +IS0°C
2 One TTL Load is defmed
as IlL =-1.6mA mIX @ VIL =O.4V, IB-I=40I1A mIX @ VB-I =2.4V.
One LSTTL Load is defined as IIL=~.36rnA
max@ VIL=O.4V,lrn=20I1A
max@ Vrn=2.7V.
3 AD362KD External Hold Capacitor is limited to +8SoC; AD362 device itself may be stored at up to +lS0°C.
subject
to change without
notice.
-2-
-
TE
8 ~
*
*
*
The AD362 is available full;' processed and screened to the requirements of MIL-STD-883, Class B. A complete
list of tests is given on page 3. When ordering, specify "AD362SD/883B".
Specifications
.t
8
~
~"
ABSOLUTE MAXIMUM RATINGS
t
;
I
+V,DigitaISupply
+5.5V
+V, Analog Supply
+16V
-V, Analog Supply
-16V
! VIN, Signal
Pin
Number
:tV, Analog Supply
i VlN, Digital
0 to +V, Digital Supply
:t1V
, AcND to DGND
,
STANDARD PROCESSING
OBS
PROCESS
CONDITIONS
1) 100% pre-cap Visual
Inspection
In-house Criteria
2) Stabilization
24 hours @ +lS0°C
Bake
I~
OLE
24 hours@+12SoC
PROCESSING TO MIL-STD-883
PROCESS
CONDITIONS
1) 100% pre-cap Visual
Inspection
2017.1
2) Stabilization
Bake
1008, 24 hours @ +lS0°C
3) Temperature
Cycle
1010, Test Condition C, 10 cycles,
-65°C to +lS0°C
4) Constant Acceleration
2001, VI Plane, 1000G
5) Visual Inspection
Visable Damage
6) Operating Burn-In
1015, Test Condition B 160 hours @
+12SoC
7) Seal Test:
1014, Test Condition A, 5 x 10-7 std cc/sec
1014, Condition C
2009
Normally Connected to ADC StatUs
Offset Adjust (See Figure 5)
Offset Adjust (See Figure 5)
Analog Output
Normally Connected to ADC
"Analog In':
Analog Ground
"High" ("Low") Analog Input, Channel
"High" ("Low") Analog Input, Channel
Negative Analog Power Supply, -15V
Positive Analog Power Supply, +15V
"High" ("Low") Analog Input, Channel
"High" ("Low") Analog Input, Channel
"High" ("Low") Analog Input, Channel
"High" ("Low") Analog Input, Channel
"High" ("Low") Analog Input, Channel
"High" ("Low") Analog Input, Channel
Input Channel Select, Address Bit AE
Input Channel Select, Address Bit AO
Input Channel Select, Address Bit Al
Input Channel Select, Address Bit A2
Input Channel Select Latch
"0": Latched
14
15
16
TE
17
18
19
20
21
22
23
24
25
26
27
28
29
30
All models of AD362 ordered to the requirements of MIL-STD-883.
Method 5008 are identified with a /883B suffix and receive the
following processing:
9) External Visual Inspection
11
12
13
i
4) Operating Bum-In
Per Data Sheet
9
I
I
Method 1014 Test Condition C
8) Final Electrical Test
Digital Ground
Positive Digital Power Supply, +5V
"High" Analog Input, Channel 7
"High" Analog Input, Channel 6
"High" Analog Input, Channel 5
"High" Analog Input, Channel 4
"High" Analog Input, Channel 3
"High" Analog Input, Channel 2
"High" Analog Input, Channell
"High" Analog Input, Channel 0
Hold Capacitor (Provided)
Sample-Hold Command
"0": Sample Mode
"1": Hold Mode
110
1
3) Seal Test, Gross Leak
Fine Leak
Gross Leak
2
8
1
As part of the standard manufacturing procedure, all models of rhe
AD362 receive the following processing:
Single-End/Differential Mode Select
"0": Single-Ended Mode
"1": Differential Mode
5
6
7
I
I
i
Function
1
3
4
I
PROCESSING FOR HIGH RELIABILITY
t
----
,
(ALL MODELS)
l
"1": Latch Transparent
AD362 ORDERING GUIDE
Model
8
AD362KD
AD362SD
AD362SD/
883B
NOTE:
D Suffix
Specification
Temp Range
Max Gain
TC
(1-24)
Prices
(25-99)
(100+)
0 to +70°C
-55°C to +125°C
-55°C to +125°C
:t4ppm/C
:t2ppm/ C
:t2ppm/C
$160.00
$295.00
$365.00
$139.50
$270.00
$330.00
$119.50
$230.00
$280.00
--= Dual-In-Line
package designator.
-3--
--
15 (7)
14 (6)
13
12
11
10
9
8
(5)
(4)
(3)
(2)
(1)
(0)
----\.0362 DESIGN
rhe AD362 consists of two 8-channel multiplexers, a differen:ial amplifier, a sample-and-hold with high-speed output buf:er, channel address latches and control logic as shown in
~igure 1. The multiplexers can be connected to the differential
lmplifier in either an 8-channel differential or 16-channel
;ingle-ended configuration. A unique fearure of the AD362 is
In internal analog switch controlled by a digital input that
Jerforms switching between single-ended and differential
nodes. This fearure allows a single AD362 to perform in
~ither mode without external hard-wire interconnections.
Of more significance is the ability to serve a mixtUre of both
;ingle-ended and differential sources with a single AD362 by
:Iynamically switching the input mode control.
smaller capacitor will allow faster sample-and-hold response
but will decrease accuracy while a larger capacitor will increase accuracy at slower conversion rates.
The output buffer is a high speed amplifier whose output
impedance remains low and constant at high frequencies.
Therefore, the AD362 may drive a fast, unbuffered, precision
ADC without loss of accuracy.
The AD362 is constructed on a substrate that includes thickfilm resistors for non-critical applications such as input protection and biasing. A separately-mounted laser-trimmed thinfilm resistor network is used to establish accurate gain and
high common-mode rejection. The metal package affords
electromagnetic and electrostatic shielding and is hermetically
welded at low temperatures. Welding eliminates the possibility
of contamination from solder particles or flux while low temperatUre sealing maintains the accuracy of the laser-trimmed
thin-film resistors.
'"'GW' A.ACOG '.PUTS
THEORY OF OPERATION
Concept
OBS
1
..'OW..
ANACOG
ONPU'
'"ANNEL
"LEO'
r
'NPUTS
,J
The AD362 is intended to be used in conjunction with a highspeed precision analog-to-digital converter to form a complete
data acquisition system (DAS) in microcircuit form. Figure 2
shows a general AD362-with-ADC DAS application.
OLE
AO362
..of
",,"
:{
Multiplexer channel address inputs are interfaced through a
level-triggered ("transparent") input register. With a Logic "1"
at the Channel Select Latch input, the address signals feed
through the register to directly select the appropriate input
channel. This address information can be held in the register
by placing a Logic "0" on the Channel Select Latch input. Inrernallogic monitors the status of the Single-Ended/Differential
lode inp<lt and addresses the multiplexers accordingly.
JI
'""""'"
~O"."'ON
,,",,""
TE
'"ANNEL
"LEO'
Figure 1. A0362 Analog Input Section Functional Block
Diagram and Pinout
""",,
,m
'"'
88
Figure 2. AD362 with ADC as a Complete Data Acquisition
System
By dividing the data acquisition task into two sections, several
important advantages are realized. Performance of each design
is optimized for its specific function. Production yields are
increased thus decreasing costs. Furthermore, the standard
configuration packages plug into standard sockets and are
easier to handle than larger packages with higher pin counts.
. differential amplifier buffers the multiplexer outputs while
roviding high input impedance in b0th differential and. singlei1ded modes. Amplifier gain and common mode rejection are
ctively laser-trimmed.
Svstem Timine:
Figure 3 is a timing diagram for the AD362 connected as
shown in Figure 2 and operating at maximum conversion rate.
The ADC is assumed to be a conventional 12 bit type such as
the AD572 or AD ADC80.
lhe sample-and-hold is a high speed monolithic device that can
also function as a gated operational amplifier. Its uncommitted
differential inputs allow it to serve a second role as the output
subtractor in the differential amplifier. This eliminates one
amplifier and decreases drift, settling time and power consumption. A Logic "1" on the Sample-and-Hold Command input
will cause the sample-and-hold to "freeze" the analog signal
while the ADC performs the conversion. Normally the Sampleand-Hold Command is connected to th~ ADC Status output
which is at Logic" 1" during conversion and Logic "0" between conversions. For slowly-changing inputs, throughput
speed may be increased by grounding the Sample-and-Hold
Command input instead of connecting it to the ADC status.
ADDRESS
ADDRESS
CONVERT
A Polystyrene hold capacitor is provided with each commercial temperatUre range device (AD362KD) while a Teflon capacitor is provided with units intended for operation at temperatures up to 125°C (AD362SD). Use of an external capacitor
allows the user to make his own speed/accuracy tradeoff; a
STATUS
~
lATCH
COMMAND
,.
{SAMPlE.HOlD)
GATED
CLOCK
Figure 3. DAS Timing Diagram
-4--
I-
-
.
.
The normal sequence of events is as follows:
1. The appropriate Channel Select Address is latched into the
address register. Time is allowed for the multiplexers to
settle.
2. A Convert Start command is issued to the ADC which, in
response, indicates that it is "busy" by placing a Logic "1"
on its Status line.
2. Trigger the oscilloscope on Status. Delay the display such
that Status is mid-screen.
3. Observe the LSB data output of the ADC.
4. Vary the analog input control to confirm that the LSB
transition precedes the Status transition.
Sin~le-Ended/Differential Mode Control
The AD362 features an internal analog switch that configures
the Analog Input Section in either a 16-channel single-ended
or 8-channel differential mode. This switch is controlled by a
TTL logic input applied to pin 1 of the Analog Input Section:
3. The ADC Status controls the sample-and-hold. When the
ADC is"busy", the sample-and-hold is in the Hold mode.
4. The ADC goes into its conversion routine. Since the sampleand-hold is holding the proper analog value, the address
may be updated during conversion. Thus multiplexer settling time can coincide with conversion and need not affect
throughput rate.
"0": Single-Ended (16 channels)
"1 ": Differential (8 channels)
When in the differential mode, a differential source may be
applied between corresponding "High" and "Low" analog
input channels.
OBS
5. The ADC indicates completion of its conversion by returning Status to Logic "0". The sample-and-hold returns to
the Sample mode.
It is possible to mix SE and DIFF inputs by using the mode
control to command the appropriate mode. In this case, four
microseconds must be allowed for the output of the Analog
Input Section to settle to within :to.OI % of its final value, but
if the mode is switched concurrent with changing the channel
address, no significant additional delay is introduced. The
effect of this delay may be eliminated by changing modes
while a conversion is in progress (with the sample-and-hold
in the "Hold" mode). When SE and DIFF signals are being
processed concurrently, the DIFF signals must be applied
between corresponding "High" and "Low" analog input channels. Another application of this feature is the capability of
measuring 16 sources individually andlor measuring differences
between pairs of those sources.
6. If the input signal has changed full-scale (different channels
may have widely-varying data) the sample-and-hold will
typically require 10 microseconds to "acquire" the next
input to sufficient accuracy for 12-bit conversion.
OLE
Mter allowing a suitable interval for the sample-and-hold to
stabilize at its new value, another Convert Start command may
be issued to the ADC.
+15V
51<"
I
g~G,'lAL
ANALOG
INPUT
20""
51<"
OUTPUTS
:
jf
I
MSB
BIT 2
LSB
ADC
UNDER
TE
TEST
Input Channel Addressin~
Table 1 is the truth table for input channel addressing in both
the single-ended and differential modes. The 16 single-ended
channels may be addressed by applying the corresponding
digital number to the four Input Chanl1el Select address bits,
AE, AD, AI, A2 (pins 28-31). In the differential mode, the
eight channels are addressed by applying the appropriate
digital code to AD, Al and A2; AE must be enabled with a
Logic "1 ". Internal logic monitors the status of the SE/DIFF
Mode input and addresses the multiplexers singularly or in
pairs as required.
-15V
STATUS
CHANNEL A "1"~
"STATUS" "o"_'_~
I
I
CHANNEL
"LSB"
'I
B
--.
I
I
,
: VALIDDATA
I
.--STATUSDELAY
Figure4. ADC Status Valid Test
NOTE:
.
Valid OutPut Data
Not all ADCs have all data bits available when Status indicates
that the conversion is complete. Successive approximation
ADCs based on the 250213/4 type of register must have a Status
delay built in or the final data bit will lag Status by approximately SOns. This will result in two problems:
1. The sample-and-hold will return to Sample, disturbing the
analog input to the ADC as it is attempting to convert the
least significant bit. This may result in an error.
2. If the falling edge of Status is being used to load the data
into a register, the least significant bit will not be valid
when loaded.
ADDRESS
AE
A2
0
0
0
0
0
0
0
0
0
1
0
1
0
1
0
1
1
0
1
0
1
0
1
0
1
1
1
1
1
1
1 --...-1
An external lOOns delay or use of an ADC with a valid Status
output is necessary to prevent this problem. The applications
shown in this data sheet ensure that all data bits will be valid.
The following test may be made to determine if the ADC
Status timing is correct:
1. Connect the ADC under test as shown in Figure 4.
A1
AO
0
0
1
1
0
0
1
1
0
0
1
1
0
0
1
1
0
1
0
1
0
1
0
1
0
1
0
1
0
1
0
!
ON ---------CHANNEL (Pin Number)
Ended
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
~
(11)
(10)
m
(8)
(7)
(6)
(5)
(4)
(27)
(26)
(25)
(24)
(23)
(22)
(19)
~-Q~l
Differential
"Hi"
"La"
0
1
2
3
4
5
6
None
None
~~
None
None
None
None
None
(11)
(10)
(9)
(8)
(7)
(6)
(5)
7 ~4)_-
0 (27)
1 (26)
2 (25)
3 (24)
5 (23)
5 (22)
6 (19)
7 (18)
Table 1. Input Channel Addressing Truth Table
-5--
---
-
.
When the channel address is changed, six microseconds must
be allowed for the Analog Input Section to settle to within
:to.Ol % of its final output (including settling times of all
elements in the signal path). The effect of this delay may be
eliminated by performing the address change while a conversion is in progress (with the sample-and-hold in the "Hold"
mode).
Inout Channel Address Latch
The AD362 is equipped with a latch for the Input Channel
Select address bits. If the Latch Control pin (pin 32) is at
Logic "1 ", input channel select address information is passed
through to the multiplexers. A Logic "0" "freezes" the input
channeraddress present at the inputs at the "1"-to-"O" transition (level-triggered).
This feature is useful when input channel address information
is provided from an address, data or control bus that may be
required to serv:ce many devices. The ability to latch an
address is helpful whenever the user has no control of when
address information may change.
OBS
Samole-and-Hold Mode Control
The Sample-and-Hold Mode Control input (pin 13) is normally
connected to the Status output (pin 20) from an analog to
digital converter. When a conversion is initiated by applying a
Convert Start command to the ADC, Status goes to Logic" 1",
putting the sample-and-hold into the "Hold" mode. This
"freezes" the information to be digitized for the period of
conversion. When the conversion is complete, Status returns
to Logic "0" and the sample-and-hold returns to the "Sample"
mode. Eighteen microseconds must be allowed for the sampleand-hold to acquire ("catch up" to) the analog input to within
:to.Ol % of the final value before a new Convert Start command is issued.
Analo!! Inout Section Offset Adjust Circuit
Although the offset voltage of the AD362 may be adjusted,
that adjustment is normally performed at the ADC. In some
special applications, however, it may be helpful to adjust the
offset of the Analog Input Section. An example of such a case
would be if the input signals were small «10mV) relative to
AD362 voltage offset and gain was to be inserted between the
AD362 and the ADC. To adjust the offset of the AD362, the
circuit shown in Figure 5 is recommended.
lold Caoacitor
\ 2000pF capacitor is provided with each AD362. One side
)f this capacitor is wired to pin 12, the other to analog ground
as close to pin 17 as possible. The capacitor provided with the
AD362KD is Polystyrene while the wider operating temperature
range of the AD362SD requires a Teflon capacitor (supplied).
OUTPUT
16
Figure 5. AD362 Offset Voltage Adjustment
Under normal conditions, all calibration is performed at the
ADC Section.
Other Considerations
Grounding: Analog and digital signal grounds should be kept
separate where possible to prevent digital signals from flowing
in the analog ground circuit and inducing spurious analog signal noise. Analog Ground (pin 17) and Digital Ground (pin 2)
are not connected internally; these pins must be connected
externally for the system to operate properly. Preferably, this
connection is made at only one point, as close to the AD362
as possible. The case is connected internally to Digital Ground
to provide good electrostatic shielding. If the grounds are not
tied common on the same card with the AD362, the digital
and analog grounds should be connected locally with back-toback general-purpose diodes as shown in Figure 6. This will
protect the AD362 from possible damage caused by voltages
in excess of :tl volt between the ground systems which could
occur if the key grounding card should be removed from the
overall system. The device will operate properly with as much
as :t200m V between grounds, however this difference will be
reflected directly as an input offset voltage.
TE
AD362
DGND
Smaller capacitors will allow slightly faster operation, but only
with increased noise and decreased precision. 1000pF will
typically allow acquisition to 0.1 % in four microseconds.
Larger capacitors may be substituted to reduce noise, and
sample-to-hold offset, but acquisition time of the sample-andhold will be extended. If less than 12 bits of accuracy is required, a smaller capacitor may be used. This will shorten the
S/H acquisition time. In all cases, the proper capacitor dielectric must be used; i.e., Polystyrene (AD326KD only) or
Teflon (AD362KD or SD). Other types of capacitors may
have higher dielectric absorption (memory) and will cause
errors. CAUTION: Polystyrene capacitors will be destroyed
if subjected to temperatures above +850 C. No capacitor is
required if the sample-and-hold is not used.
.
8
AD362
ANALOG
INPUT
SECTION
OLE
The purpose of a sample-and-hold is to "stop" fast changing
input signals long enough to be converted. In this application,
it also allows the user to change channels and/or SEt DIFF
mode while a conversion is in progress thus eliminating the
effects of multiplexer, analog switch and differential amplifier
settling times. If maximum throughput rate is required for
slowly changing signals, the Sample-and-Hold Mode Control
may be wired to ground (Logic "0") rather than to ADC
,tatus thus leaving the sample-and-hold in a continuous
;ample mode.
1
88
ADC
AGND
TO
CARD
CONNECTOR
Figure 6. Ground-Fault Protection Diodes
Power Supply Bypassing: The :t15V and +5V power leads
should be capacitively bypassed to Analog Ground and Digital
Ground respectivelyfor optimum device performance. IJ.LF
tantalum types are recommended; these capacitors should be
located close to the system. It is not necessary to shunt these
capacitors with disc capacitors to provide additional high
frequency power supply decoupling since each power lead is
bypassed internally with a 0.039J.LFceramic capacitor.
-6-
,
.
-
""
~,:~:2~~-~!:.~
:~~,~
~
-
~
"'~f'::
Interfacing to Popular Analog to Digital Converters
The AD362 has been designed to interface directly to most
analog to digital converters; often no additional components
are required and only two interconnections must be made.
The direct interface requirements for the ADC are as follows:
Figure 7a shows the AD362 driving an AD ADC80. The
AD ADC80 is a 12-bit, 25-microsecond, low-cost ADC that
meets all of the requirements listed above. Throughput rate
is typically 30kHz with no missing codes over the operating
temperatUre range.
Figure 7b shows a lO-bit application based on the AD362 and
the AD571, a complete low cost 10-bit, 25-microsecond ADC.
In this case, two of the above requirements are not met:
1. The ADC StatUs output must be positive-true Logic ("1"
during conversion).
2. Transition from "0" to "1" must occur at least 200ns
before the most significant bit decision is made (successive
approximation ADC) or before input integration starts
(integrating type ADC).
3. StatUs must not retUrn to "0" before the LSB decision is
made (see page 5).
4. If StatUs is being used to latch output data, it must not
retUrn to Logic "0" until all output data bits are valid and
available.
1. DR (DATA READY), as StatUs, is positive-true but. . .
2. DR does not indicate that a conversion is in progress until
1.5JJs after conversion starts.
3. DR does indicate conversion complete after the LSB decision is made, but. . .
4. DR precedes the enabling of the AD571 output 3-state
gates by 500ns.
OBS
The gating provided by Ul allows the applied convert com-mand (CC) to initiate input hold at the AD362. CC must last
for more than 1.5JJs so that DR may then assume control
of Hold. If conversion is continuous (consistent with multichannel operation), the next convert command can be used to
load the previously-converted data into an output register. For
single conversion operation, a IJJs delay of the falling edge of
DR may be used to signify valid data.
Complete system throughput performance is determined by
combining the worst-case specifications of the AD362 and the
ADC. If guaranteed system performance is required, the
AD363 and AD364 are recommended. The AD363 includes
an AD362 and an AD572 12-bit, 25-microsecond precision
ADC. The AD364 consists of an AD362 and an AD574 12-bit,
microprocessor-compatible, low cost ADC. Each is specified
as a complete, two-package system; data sheets are available
upon request.
-
OLE
DC POWER
TE
DC POWER
ANALOG
INPUTS
(IS)
CAPACITOR
ANALOG
INPUTS
1161
AOJ62
IANALOGHOLD
ANALOGI ADAOC80
OUT
IN
SAMPLE/HOLD
CHANNEl
SELECT
LATCH
4
I
0
AO571
DR
BITS
OUT
(12)
DATA
Me'
CHANNEL
SELECT
LATCH
I STATUSOUTPUT
1/474LSJ2
CONVERT
COMMAND
.n. DATA
STROBE
ITO OUTPUT
CONVERT
START
DATA
BITS
OUT
(101
~1.5~.
--::J C
REGISTERI
MIN
DATA STROBE
(TO OUTPUT
REGISTER)
b. la-Bit DAS UsingAD362 and AD571
a. l2-Bit DAS Using AD362 and AD ADC80
Figure 7. Data Acquisition Systems Based on the AD362 and Popular ADC's
I
-7-
The HAS series of ultra-fast ADCs are 8-bit (HAS080l), 10bit (HASI00l) and 12-bit (HASI202) devices that convert in
I.S, 1.7, and 2.8 microseconds (maximum) respectively. These
devices are hybrid IC's, packaged in 32-pin DIPs. Since the
Data Ready signal from the HAS precedes the LSB decision,
DR must be delayed. Figure 8b shows the appropriate circuitry to provide that delay. Throughput rate for the 12-bit
system is typically 80kHz.
Interfacing to Special Purpose ADCs
The ADS200 series of ADCs perform a 12-bit conversion in SO
microseconds and feature totally adjustment-free operation,
high accuracy, and a small hermetically-sealed 24-pin package.
These ADCs are often used in high-reliability applications
and, like the AD362SD (which operates over the -SSoC to
+12SoC temperature range) are available processed to MILSTD-883, Class B. The ADS200 series meets all of the interfacing requirements for direct connection to the AD362 as
shown in Figure 8a. System throughput rate is typically 16kHz.
8
DC POWER
0
CXI
.....
M
I
C)
I
N
18
u
DC POWER
DATA
BITS
OUT
18. 10 OR 12)
HOLD
ANALOG
INPUTS
116)
AD5200
SERIES
STATUS
CONVERT
START
a. 12-Bit High Accuracy and Reliability
AD362 and AD5200
DATA STROBE
ITO
OUTPUT
REGISTER)
DAS Using
Figure B. Data Acquisition
HAS
SERIES
ANALOG
IN
DR
SAMPLE/HOLD
ENCODE
INPUT
r
CHANNEL
C~:L~~~L
SL~L,~g
4,
---0
1
14) r---J
I
OLE
CONVERT
COMMAND
.~150ps
--:J
I
I T
C
MIN
I
~-
rnn__~__n_n_n;
1.00 125.4)
MIN
T
t
T
~
0.0812.01
DIA.TYP-I
-(-L
T
(~i,
C; 000
00
0000
..1
0000
0
0
GLASS BEAD
STANDOFFS
--J
MAX
f-I
0
I
'THIS DIMENSION IS FOR POL YSTYRENE CAPACITOR
SUPPLIED WITH AD362KD
0
0
0 00
L
,-
00
00
0000000
0.100. TYP
12.51
0 0
T
1
I~~)
~
CI!
::>
z
0
w
IZ
a:
a..
0.025
10.46.0.05)
0.018
!T0.00~
0.2015.11
DIA.
YP
(~662)
-L
0.35
18.9)
--I
}
DATE CODE
I
I.
PIN 1,
8
MAX BODY LENGTH OF TEFLON CAPACITOR SUPPLIED
WITH AD362SD IS 1.00".
GREEN~"""
GLASS BEAD
MAX
I
.I.
~
I
":'...J
0.69 117.5)
IKD OR SO)
.8
0211531
:
HOLD CAPACITOR
PIN 1./
INDENTIFIER
--1-
J
TE
shown in inches and (mm).
MODEL DESIGNATION
-
1-
I
I
I
b. High-Speed DAS Using AD362 and HAS
"D" PACKAGE
C==J
1oopF
U1:7400
OUTLINE DIMENSIONS
PACKAGE SPECIFICATIONS
CAD362
'c",
- - ~~v-.:'~~T
Systems Based on the AD362 and Purpose ADCs
Dimensions
DATA
STROBE
(TO OUTPUT
REGISTER)
COMMAND
OUTPUT
Jl
CLOCK
INPUT
CAPACITOR
ANALOG
OUT
DATA
BITS
OUT
112)
OBS
ANALOG
INPUTS
(161
7
AD362
-- t
10.6)
- ~-L
1
0.900
1
122.9)-1
METAL 32 PIN DUAL.IN.LINE
PACKAGE
NOTES:
11 PACKAGE: KOVAR WITH 100pIN. MIN., NICKEL PLATE
2) PINS: KOVAR WITH 50p1N. MIN 24K GOLD PLATE
3) PACKAGE AND PINS MEET ALL REQUIREMENTS OF MIL-STD-883
41 TOLERANCES, UNLESS OTHERWISE NOTED:
al .XX: '.01 1.25)
h) .XXX: '.005 1.13)
MATING SOCKET:
ROBINSDN.NUGENT PART NUMBER SB.16.G
12 REQUIRED)
ROBINSON.NUGENT,
INC.
800 EAST EIGHTH STREET
NEW ALBANY, IN 47150
1812) 945-0211
-8-
8