www.fairchildsemi.com
FM50
Analog Temperature Sensor
Features
•
•
•
•
•
•
•
Analog output: 10 mV/°C
Range: -40 to 125°C range:
Accuracy: ±2°C at 25°C
Supply Current, 170 µA max.
Output Drive, 25 µA
Self-heating < 0.1°C
Operating Voltage: +2.4V to +6V
Description
As a precision CMOS temperature sensor, the FM50 is
cost-effective for accurate low-power temperature monitor-
ing applications. Output voltage versus temperature is
extremely linear. With no load, the supply current is
typically 130µA. For normal operation, the load on V
OUT
should be 100K
Ω
or less.
In a typical application, a remotely mounted FM50 is
monitored by a microcontroller with an analog A/D
converter input. Alternatively, the FM50 can drive a
comparator with a high impedance input.
Accuracy is typically ± 0.5°C at room temperature; and
better than ±2°C from 0 to 75°C.
Available in both standard and Lead Free 3-pin SOT-23
package.
Applications
•
•
•
•
•
•
•
•
Mobile Phones
Computers
Battery Management
Office Equipment
HVAC
Power Supply Modules
Disk Drives
Automotive
Thermal Response
V
OUT
(mV)
1750
750
500
100
-50 -40
-25
0
25
50
75
100
125
Temperature ( C)
Temperature ( C) =
V
OUT
- 500 mV
10 mV/ C
FM50 Output Voltage vs. Temperature
REV. 1.1.0 10/5/04
FM50
PRODUCT SPECIFICATION
Pin Assignments
GND
3
FM50
1
2
V
DD
V
OUT
Pin Descriptions
Pin Name Pin No.
V
OUT
V
DD
GND
2
1
3
Type
Analog Output
Power
Power
Function
Temperature Sense.
Analog output voltage indicating temperature.
V
OUT
= 500+10T(°C) mV
Supply Voltage.
2.4 to 6.0V
Ground.
2
REV. 1.1.0 10/5/04
PRODUCT SPECIFICATION
FM50
Absolute Maximum Ratings
1
Parameter
Supply Voltage
Output Voltage
Output Current
Operating Temperature Range
2
Storage Temperature Range
Lead Soldering Temperature
ESD
2
Human Body Model
Machine Model
-50
-60
Min.
Typ.
Max.
+7
V
DD
+ 0.5
10
+150
+150
220
2000
250
Units
V
V
µA
°C
°C
°C
V
V
Notes:
1. Absolute maximum ratings are limits beyond which operation may cause permanent damage to the device. These are stress
ratings only; functional operation at or above these limits is not implied.
2. Operating ratings are conditions for which the device is intended to be functional without specific guaranteed performance
limits. For guaranteed specifications and test conditions refer to Electrical Characteristics.
3. Human Body Model: 100pF capacitor discharged through a 1.5kOhm resistor into each pin. Machine Model: 200pF capacitor
discharged directly into each pin.
Electrical Characteristics
4
Limits apply for -40°C
≤
T
A
≤
+125°C and V
DD
= +5.0V unless otherwise noted.
Parameter
Transfer Characteristic
Sensitivity
Output at 25°C
Accuracy
5
Symbol
A
OUT
T
A
= +25°C
T
A
= -40°C (T
MIN
)
T
A
= +125°C (T
MAX
)
Conditions
Min
9.7
740
-2
-3.5
-3
-40
-0.8
100
I
ONSN
I
ONSG
I
OL
100k
Ω
load at 25°C
1
Sensing
Surge
25
50
10
Typ
10
750
±0.5
±1
±1
Max
10.3
760
+2
+3.5
+3
+125
+0.8
1750
Units
mV/°C
mV
°C
°C
°C
°C
°C
mV
µA
mA
µA
k
Ω
mV/mA
V
µA
°C/W
Temperature Range
Non-Linearity
6
Output
Output Voltage Range
Output Current Source
Output Current Sink
Output resistance
Load regulation
Power
Supply Voltage
Quiescent Current
Package
Thermal Resistance
V
DD
I
S
2.4
No output load
260
6.0
170
Notes:
4. These specifications are guaranteed only for the test conditions listed.
5. Accuracy (expressed in °C) = Difference between calculated output voltage and measured output voltage. Calculated output
voltage = 10mV/°C multiplied by the device’s case temperature at specified conditions of temperature, voltage and power
supply, added to 500mV.
6. Non-linearity is defined as the deviation of the output-voltage-versus-temperature curve from the best-fit straight line, over the
device’s rated temperature range.
REV. 1.1.0 10/5/04
3
FM50
PRODUCT SPECIFICATION
Typical Performance Characteristics
4.0
3.0
2.0
upper spec limit
V
DD
= +5V
Accuracy (°C)
1.0
0.0
-1.0
-2.0
-3.0
lower spec limit
-4.0
-50
0
50
100
150
Temperature (°C)
Figure 1. Accuracy vs. Temperature
VOUT Deviation from 5V Value (mV)
140.0
120.0
5.0
4.0
3.0
2.0
1.0
0.0
-1.0
2.5
Supply Current (uA)
100.0
80.0
60.0
40.0
20.0
0.0
-50
-30
-10
10
30
50
70
90
110
130
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Temperature (°C)
VDD Volts
Figure 2. Typical IDD versus Temperature
Figure 3. Typical Sensitivity to Supply Voltage
Description
Within the FM50 are a thermal diode, calibration circuits
and amplifiers. Since the FM50 is calibrated at 33°C,
the nominal output in mV is:
V
OUT
=
830
+
10
• (
T
–
33
)
These values accommodate the specified accuracies at -40,
25 and +125°C.
Output structure of the FM50 is an n-channel CMOS transis-
tor driving a p-channel load. Available current is typically 50
µA to ground. Series resistance is typically 7 k
Ω
, charging
and 2 k
Ω
, discharging through a capacitor connected from
V
OUT
to ground.
Following application of power to V
DD
, V
OUT
is accurate
following a delay of approximately 80 msec.
where T is the thermal junction temperature expressed in °C.
At 33°C, the tolerances are as follows:
1.
2.
Offset is ±3mV
Slope, ±0.3mV/°C
4
REV. 1.1.0 10/5/04
PRODUCT SPECIFICATION
FM50
Applications Information
Although the FM50 is a simple device, care must be taken to
ensure that temperature is measured accurately. There are
two major sources of errors:
3.
4.
5.
Voltage errors.
Thermal Delay Errors.
Location errors
Location Errors
Position is another source of error. Even within a controlled
thermal environment, changing location by a few inches can
lead to errors of several tenths of 1°C
Mounting
The FM50 can be easily mounted by gluing or cementing it
to a surface. In this case, its temperature will be within about
0.2°C of the temperature of the surface it is attached to if the
ambient air temperature is almost the same as the surface
temperature. If the air temperature is much higher or lower
than the surface temperature, the actual temperature of the
FM50 die will be at an intermediate temperature between the
surface temperature and the air temperature.
Voltage Errors
VDD
FM50
VOUT
GND(power)
GND(sense)
Figure 4. Recommended Electrical Connections
To ensure good thermal conductivity, the backside of the
FM50 die is directly attached to the GND pin. The lands and
traces to the FM50 will, of course, be part of the printed
circuit board, which is the object whose temperature is being
measured. These printed circuit board lands and traces will
not cause the FM50’s temperature to deviate from the
desired temperature.
Alternatively, the FM50 can be mounted inside a sealed-end
metal tube, and can then be dipped into a bath or screwed
into a threaded hole in a tank. As with any IC, the FM50 and
accompanying wiring and circuits must be kept insulated and
dry to avoid leakage and corrosion. This is especially true if
the circuit may operate at cold temperatures where conden-
sation can occur. Printed-circuit coatings and varnishes such
as Humiseal and epoxy paint or dips can be used to ensure
that moisture cannot corrode the FM50 or its connections.
A Kelvin connection is recommended to avoid errors due to
voltage drops in the ground connections. Although the
typical 130µA supply current draw of the FM50 will only
cause a 130µV error if the series resistance is 1
Ω
, a 100 mA
current supply to adjacent circuits can cause a 10mV drop
across 100m
Ω
(10m
Ω
is a typical value for soldered joints or
contact resistance), leading to a 1°C error. For this reason,
the FM50 should be Kelvin connected as shown in Figure 4.
Thermal Delay Errors
For measurement accuracy of the order of tenths of 1°C,
adequate settling time must be allowed. For a typical circuit
board installation, 15 minutes should be allowed to elapse
following reading of temperature within 1 - 2°C of the
expected final value. Once V
OUT
has ceased to slew and
is stable (with or without about ±0.1°C noise) for about 5
minutes, temperature can be calculated.
REV. 1.1.0 10/5/04
5
京公网安备 11010802033920号
Copyright © 2005-2026 EEWORLD.com.cn, Inc. All rights reserved
电子工程世界
