MAX1153/MAX1154
Stand-Alone, 10-Channel, 10-Bit System Monitors
with Internal Temperature Sensor and V
DD
Monitor
______________________________________________________________________________________ 25
Manual conversion mode
External reference for voltage measurements
Channel Enable Register Power-Up/Reset Defaults
At power-on or after a RESET command, the channel
enable register resets to FF hex, enabling all channels:
the internal temperature sensor, the V
DD
monitor, and
AIN0–AIN7.
Input Configuration Register
Power-Up/Reset Defaults
At power-on or after a RESET command, the input con-
figuration register resets to 00 hex, configuring
AIN0–AIN7 for single-ended voltage measurement.
Alarm Register Power-Up/Reset Defaults
At power-on or after a RESET command, the alarm reg-
ister is reset to 000000 hex, indicating that no alarm
condition exists.
Current Data Register Power-Up/Reset Defaults
At power-on or after a RESET command, each chan-
nel’s current data register is reset to 200 hex.
Upper Threshold Register Power-Up/Reset Defaults
At power-on or after a RESET command, each chan-
nel's upper threshold register is reset to 3FF hex. This
state effectively disables the upper threshold.
Lower Threshold Register Power-Up/Reset Defaults
At power-on or after a RESET command, each chan-
nel's lower threshold register is reset to 000 hex. This
state effectively disables the lower threshold.
Channel Configuration Register
Power-Up/Reset Defaults
At power-on or after a RESET command, each chan-
nel's configuration register is reset to 000 hex, which
configures the fault bits to cause an alarm to occur on
the first overrange or underrange condition and dis-
ables averaging.
Computing Data Throughput
The MAX1153/MAX1154 throughput rate depends on
the number of enabled channels, their configuration
(temperature or voltage), and the reference mode.
Voltage measurements require 10.6µs (typ) to com-
plete, and temperature measurements require 46µs.
Channel pairs configured for differential measurements
count as only one for throughput computation.
The reference system takes 20µs to power up in refer-
ence mode 00 prior to each temperature measurement,
40µs to power up in reference mode 01 after each sam-
ple wait period (if sample wait time > 80µs), and no
power-up time in reference mode 10.
The sampling period is calculated as follows:
t
sw
= (t
pu
) + (N
v
)t
conv[volt]
+ (N
t
)t
conv[temp]
+ t
wait
where:
t
sw
= all channels scan sampling period
t
pu
= reference power-up time
t
conv[volt]
= voltage-configured channel conversion time
N
v
= number of voltage-configured channels
t
conv[temp]
= temperature-configured channel conver-
sion time
N
t
= number of temperature-configured channels
t
wait
= sample wait time
The terms in the previous equation are determined as
shown by the number of enabled channels, the input
channel configuration (voltage vs. temperature), the
sample wait time, and the reference mode. The follow-
ing calculation shows a numeric example:
t
sw
= 40µs + 8 x 10.6µs + 2 x 46µs + 395µs = 611.8µs
40µs is the time required for the reference to power-
up (reference mode = 00) every time the
MAX1153/MAX1154 come out of automatic shut-
down mode after a sample wait period.
8 x 10.6µs is the time required for seven channels
configured for voltage measurement and the V
DD
monitor.
2 x 46µs is the time required for temperature mea-
surement (46µs for each temperature measurement
(internal or external)).
395µs is the sample wait time, set by bits B5, B6,
B7 of the setup register (see Tables 7 and 8).
The MAX1153/MAX1154 use an internal clock for all
conversions. The serial interface clock does not affect
conversion time.
Performing eight single-ended remote channels tem-
perature measurements, an internal temperature mea-
surement, and an internal V
DD
measurement with a
sample wait time of zero results in an average conver-
sion rate of 24ksps or 2.4ksps per channel.
Performing eight single-ended voltage measurements,
an internal temperature measurement, and an internal
V
DD
measurement with sample wait time of zero results
in an average conversion rate of 70ksps or 7ksps per
channel.
MAX1153/MAX1154
Stand-Alone, 10-Channel, 10-Bit System Monitors
with Internal Temperature Sensor and V
DD
Monitor
26 ______________________________________________________________________________________
Automatic Reference Shutdown
The MAX1153/MAX1154 enter an automatic shutdown
mode when in reference mode 00 or when the sample
wait is greater than 80µs in reference mode 01. Using
either of these reference modes and a sample wait
period as long as the application allows results in the
lowest power consumption.
Temperature Measurement
The MAX1153/MAX1154 support both single-ended
and differential temperature measurements. The design
decision between the two types of measurements
depends on the desired level of accuracy and on type
and/or number of temperature sensors. The superior
common-mode rejection and lower noise of the differ-
ential mode reduces measurement errors and provides
higher accuracy, while single-ended measurements
require a lower number of connections, resulting in a
simpler implementation and a higher number of moni-
tored points for each MAX1153/MAX1154.
Differential Temperature Measurement
Connect the anode of a diode-connected transistor to
the even input channel and the cathode to the odd
input channel of an input pair configured for differential
temperature measurement (AIN0/AIN1, AIN2/AIN3,
AIN4/AIN5, or AIN6/AIN7). Run the two sensor connec-
tion lines parallel to each other with minimum spacing.
This improves temperature measurement accuracy by
minimizing the differential noise between the two lines,
since they have equal exposure to most sources of
noise. For further improved noise rejection, shield the
two sensor connections by running them between
ground planes, when available.
Configure the MAX1153/MAX1154 inputs for differential
temperature measurement in the input configuration
register (see Tables 9 and 10) and enable the even
channel number in the channel enable register (see
Table 4).
Single-Ended Temperature Measurement
Connect the anode of a diode-connected transistor to
the input channel and the cathode to ground. Choose
ground connections for sensors away from high-current
return paths to avoid the introduction of errors caused by
voltage drops in the board/system ground, which is the
main drawback for single-ended measurements.
Practical options for better accuracy are the use of a
star-configured subsystem ground or a signal ground
plane; to isolate the anode sensor connection trace away
from board and system noise sources; or to shield it with
ground lines and ground planes (when available) to pre-
vent accuracy degradation in the temperature measure-
ments caused by magnetic/electric noise induction.
Configure the MAX1153/MAX1154 input used for single-
ended temperature measurement in the input configura-
tion register (see Tables 9 and 10) and enable the
analog input in the channel-enable register (see Table 4).
Remote Temperature Sensor Selection
Temperature-sensing accuracy depends on having a
good-quality, diode-connected, small-signal transistor
as a sensor. Accuracy has been experimentally verified
for 2N3904-type devices. The transistor must be a
small-signal type with low base resistance. Tight speci-
fications for forward current gain (+50 to +150, for
example) indicate that the manufacturer has good
process controls and that the devices have consistent
V
BE
characteristics. CPU on-board sensors and other
ICs’ on-board temperature-sensing devices can also
be used (see Table 16 for recommended devices).
Table 16. Remote Sensor Transistor
Manufacturers
MANUFACTURER MODEL NUMBER
Central Semiconductor (USA) CMPT3904
Fairchild Semiconductors (USA) MMBT3904
Motorola (USA) MMBT3904
Rohm Semiconductor (Japan) SST3904
Siemens (Germany) SMB3904
Zetex (England) FMMT3904CT-ND
Diodes Inc. MMBT3904
OUTPUT CODE
FULL-SCALE
TRANSITION
11....111
0
0
INPUT VOLTAGE (LSB)
ZS = 0
FS = V
REF
11....110
11....101
00....011
00....010
00....001
00....000
213 FS
1 LSB =
V
REF
1024
FS = 3/2 LSB
Figure 10. Unipolar Transfer Function, Full Scale (FS) = V
REF
MAX1153/MAX1154
Stand-Alone, 10-Channel, 10-Bit System Monitors
with Internal Temperature Sensor and V
DD
Monitor
______________________________________________________________________________________ 27
Transfer Function
Figure 10 shows the nominal transfer function for sin-
gle-ended or differential unipolar configured inputs,
Figure 11 illustrates the transfer function for differential
bipolar conversions, and Figure 12 shows temperature
conversions. Code transitions occur halfway between
successive-integer LSB values. Output coding is bina-
ry, with 1 LSB = 2.44mV (MAX1153) or 4mV (MAX1154)
for unipolar and bipolar operation, and 1 LSB = +0.5°C
(MAX1153/MAX1154) for temperature measurements.
For unipolar operation, the 0 code level transition is at
[1/2(V
REF
/ 1024)].
The FFF hex level transition is at [1022.5(V
REF
/ 1024)].
1 LSB = V
REF
/ 1024.
Layout, Grounding, and Bypassing
For best performance, use PC boards. Do not use wire-
wrap boards. Board layout should ensure that digital
and analog signal lines are separated from each other.
Do not run analog and digital (especially clock) signals
parallel to one another or run digital lines underneath
the MAX1153/MAX1154 package. High-frequency
noise in the V
DD
power supply can affect the
MAX1153/MAX1154 performance. Bypass the V
DD
sup-
ply with a 0.1µF capacitor from V
DD
to GND close to
the V
DD
pin. Minimize capacitor lead lengths for best
supply-noise rejection. If the power supply is very
noisy, connect a 10Ω resistor in series with the supply
to improve power-supply filtering.
Definitions
Integral Nonlinearity
Integral nonlinearity is the deviation of the values on the
actual transfer function from a straight line. This straight
line can be either a best-straight-line fit or a line drawn
between the end points of the transfer function, once off-
set and gain errors have been corrected. The static lineari-
ty parameters for the MAX1153/MAX1154 are measured
using the end-point-fit method. INL is specified as the
maximum deviation in LSBs.
Differential Nonlinearity (DNL)
Differential nonlinearity is the difference between an
actual step width and the ideal value of 1 LSB. A DNL
error specification of less than 1 LSB guarantees no
missing codes and a monotonic transfer function.
Offset Error
The offset error is the difference between the ideal and
the actual analog input value at the first transition of the
ADC, usually from digital code 0 to code 1 for straight
binary output. For the MAX1153/MAX1154, the transi-
tion between code 0 and code 1 should occur at an
input voltage of 1/2 LSB, or 1.22mV for the MAX1153
and 2mV for the MAX1154.
OUTPUT CODE
011....111
TEMPERATURE °C
011....110
000....001
111....101
100....001
100....000
111....111
111....110
000....000
0
000....010
-256°C +255.5°C
Figure 12. Temperature Transfer Function
OUTPUT CODE
011....111
INPUT VOLTAGE (LSB)
011....110
000....001
111....101
100....001
100....000
111....111
111....110
000....000
0
000....010
-FS +FS - 1 LSB
FS =
ZS = 0
V
REF
2
-FS =
-V
REF
2
1 LSB =
V
REF
1024
Figure 11. Bipolar Transfer Function, Full Scale (±FS) = ±V
REF
/2

MAX1153BEUE+T

Mfr. #:
Manufacturer:
Maxim Integrated
Description:
Analog to Digital Converters - ADC Stnd-Alne 10Ch 10Bit System Monitor
Lifecycle:
New from this manufacturer.
Delivery:
DHL FedEx Ups TNT EMS
Payment:
T/T Paypal Visa MoneyGram Western Union

Products related to this Datasheet