ETC TGS2442

PRODUCT INFORMATION
TGS 2442 - for the detection of Carbon Monoxide
Features:
Applications:
* Low power consumption
* High sensitivity/selectivity to
carbon monoxide (CO)
* Miniature size
* Low sensitivity to alcohol vapor
* Long life and low cost
* Low humidity dependency
* CO detectors
* Air quality controllers
* Indoor parking lot ventilation
TGS 2442 utilizes a multilayer sensor structure. A glass layer for thermal insulation
is printed between a ruthenium oxide (RuO2) heater and an alumina substrate. A
pair of Au electrodes for the heater are formed on a thermal insulator. The gas sensing
layer, which is formed of tin dioxide (SnO2), is printed on an electrical insulation layer
which covers the heater. A pair of Au electrodes for measuring sensor resistance
are formed on the electrical insulator. Activated charcoal is filled between the internal
cover and the outer cover for the purpose of reducing the influence of noise gases.
TGS 2442 displays good selectivity to carbon monoxide, making it ideal for CO
monitors. In the presence of CO, the sensor's conductivity increases depending on
the gas concentration in the air. A simple pulsed electrical circuit operating on a one
second circuit voltage cycle can convert the change in conductivity to an output signal
which corresponds to gas concentration.
The figure below represents typical sensitivity characteristics,
all data having been gathered at standard test conditions (see
reverse side of this sheet). The Y-axis is indicated as sensor
resistance ratio (Rs/Ro) which is defined as follows:
Rs = Sensor resistance of displayed gases at
various concentrations
Ro = Sensor resistance in 100ppm CO
The figure below represents typical temperature and humidity
dependency characteristics. Again, the Y-axis is indicated as
sensor resistance ratio (Rs/Ro), defined as follows:
Rs = Sensor resistance at 30ppm, 100ppm and 300ppm
of CO at various temperatures and 50%R.H.
Ro = Sensor resistance at 300ppm of CO
at 25°C and 50% R.H.
Sensitivity Characteristics:
100
Air
Temperature/Humidity Dependency:
1000
Ethanol
10
50%RH
100
Rs/Ro
Rs/Ro
H2
1
Air
10
CO 30ppm
0.1
1
CO
CO 100ppm
CO 300ppm
0.01
1
10
100
1000
Gas Concentration (ppm)
10000
0.1
-20
-10
0
10
20
30
40
50
60
Temperature (˚C)
IMPORTANT NOTE: OPERATING CONDITIONS IN WHICH FIGARO SENSORS ARE USED WILL VARY WITH EACH CUSTOMER’S SPECIFIC APPLICATIONS. FIGARO STRONGLY
RECOMMENDS CONSULTING OUR TECHNICAL STAFF BEFORE DEPLOYING FIGARO SENSORS IN YOUR APPLICATION AND, IN PARTICULAR, WHEN CUSTOMER’S TARGET GASES
ARE NOT LISTED HEREIN. FIGARO CANNOT ASSUME ANY RESPONSIBILITY FOR ANY USE OF ITS SENSORS IN A PRODUCT OR APPLICATION FOR WHICH SENSOR HAS NOT BEEN
SPECIFICALLY TESTED BY FIGARO.
Basic Measuring Circuit:
Circuit voltage (VC) is applied across the
sensing element which has a resistance (Rs)
between the sensor’s two electrodes (pins
No. 2 and No. 3) and a load resistor (RL)
connected in series. The sensing element
is heated by the heater which is connected
to pins No. 1 and No. 4.
Heating cycle --The sensor requires
application of a 1 second heating cycle which
is used in connection with a circuit voltage
cycle of 1 second. Each V H cycle is
comprised by 4.8V being applied to the
heater for the first 14ms, followed by 0V
pulse for the remaining 986ms. The Vc cycle
consists of 0V applied for 995ms, followed
by 5.0V for 5ms. For achieving optimal
sensing characteristics, the sensor's signal
should be measured after the midpoint of
the 5ms Vc pulse of 5.0V (for reference, see
timing chart below).
NOTE: Application of a Vc pulse condition
is required to prevent possible migration of
heater materials into the sensing element
material. Under extreme conditions of high
humidity and temperature, a constant Vc
condition could result in such migration and
cause long term drift of Rs to higher values.
A 5ms Vc pulse results in significantly less
driving force for migration than a constant
Vc condition, rendering the possibility of
migration negligibly small.
Structure and Dimensions:
unit: mm
Top View
Preliminary Specifications:
Model number
TGS 2442
Sensing element type
M1
Standard package
TO-5 metal can
Target gases
Carbon monoxide
Typical detection range
Standard circuit
conditions
Electrical characteristics
under standard test
conditions
30 ~ 1000 ppm
Heater voltage cycle
VH
Circuit voltage cycle
VC
VC=0V for 995ms,
VC=5.0V±0.2V DC for 5ms
Load resistance
RL
variable (≥10kΩ)
Heater resistance
RH
17 ± 2.5Ω at room temp.
Heater current
IH
approx. 203mA(in case of VHH)
Heater power
consumption
PH
approx. 14mW (ave.)
Sensor resistance
Rs
6.81 kΩ ~ 68.1 kΩ in 100ppm of
carbon monoxide
Sensitivity
(change ratio of Rs)
β
0.23 ~ 0.49
Vout
- RL
The value of sensitivity (β) is calculated with two measured
values of Rs as follows:
β = Rs (CO,300ppm)
Rs (CO,100ppm)
ø6.0+0,-0.3
12.7±0.5
0.2
1.5
ø0.55±0.05
45˚
Carbon monoxide in air
at 20±2˚C, 65±5%RH
Same as Std. Circuit Condition
(above)
Circuit conditions
Sensor resistance (Rs) is calculated with a measured value of
Vout as follows:
5.9
10.0±1.0
Conditioning period before
test
Rs =
ø8.1±0.2
ø0.3
VHH=4.8V±0.2V DC, 14ms
VHL=0V, 986ms
Test gas conditions
Standard test conditions
ø9.2±0.2
A: 0.23 ~ 0.34
B: 0.26 ~ 0.37
C: 0.29 ~ 0.40
D: 0.32 ~ 0.43
E: 0.35 ~ 0.46
F: 0.38 ~ 0.49
The above six classification will be further subdivided into the
following rankings of Rs values in 100ppm of CO:
Code:
1
3
2
> 2 days (under review)
To facilitate usage of this sensor, TGS2442 is shipped in
presorted groupings which have a more narrowly defined
range of β:
Code:
4
1: 6.81 ~ 21.5kΩ
2: 10.0 ~ 31.6kΩ
3: 14.7 ~ 46.6kΩ
4: 21.5 ~ 68.1kΩ
ø5.1±0.1
Bottom view
an ISO9001 company
TECHNICAL INFORMATION FOR TGS2442
Technical Information for Carbon Monoxide Sensors
The Figaro TGS2442 sensor is a new type
thick film metal oxide semiconductor,
screen printed sensor which offers
miniaturization and utilizes pulse heating
for achieving low power consumption.
The TGS2442 displays high selectivity to
carbon monoxide together with improved
humidity dependency and durability.
Specifications
Page
Features..................................................................................................2
Applications...............................................................................................2
Structure...........................................................................................2
Basic Measuring Circuit...........................................................................2
Circuit & Operating Conditions...............................................................3
Mechanical Strength..............................................................................3
Dimensions...................................................................................................3
Operation Principle.........................................................................................................4
Basic Sensitivity Characteristics
Sensitivity to Various Gases............................................................5
Temperature and Humidity Dependency.............................................5
Gas Response Pattern.................................................................................6
Heater Voltage Dependency...........................................................................6
Initial Action.........................................................................................7
Influence of Unenergized Storage...............................................................7
Reliability
Interference Gas Test......................................................................................8
Long-Term Stability................................................................................9
Corrosion Test...........................................................................................9
Variable Ambient Temperature Test...............................................................9
Humidity Test.............................................................................................10
Stability Tests...................................................................................................10
Circuit Examples
Basic Circuit Including Trouble Detection...............................................11
Calibration and Temperature Compensation.............................................11
Marking and Packaging..................................................................................................13
See also Technical Brochure “Technical Information on Usage of TGS Gas Sensors
for Explosive/Toxic Gas Alarming”.
IMPORTANT NOTE: OPERATING CONDITIONS IN WHICH FIGARO SENSORS ARE USED WILL VARY
WITH EACH CUSTOMER’S SPECIFIC APPLICATIONS. FIGARO STRONGLY RECOMMENDS
CONSULTING OUR TECHNICAL STAFF BEFORE DEPLOYING FIGARO SENSORS IN YOUR APPLICATION
AND, IN PARTICULAR, WHEN CUSTOMER’S TARGET GASES ARE NOT LISTED HEREIN. FIGARO
CANNOT ASSUME ANY RESPONSIBILITY FOR ANY USE OF ITS SENSORS IN A PRODUCT OR
APPLICATION FOR WHICH SENSOR HAS NOT BEEN SPECIFICALLY TESTED BY FIGARO.
Revised 03/00
1
TECHNICAL INFORMATION FOR TGS2442
1. Specifications
1-1
Features
Non woven fabric
* Miniature size and low power consumption
* High sensitivity/selectivity to carbon monoxide
(CO)
* Low sensitivity to alcohol vapor
* Reduced influence by various interference gases
* Long life and low cost
Metal cap
Charcoal filter
Metal gauze (double layer)
Lead wire
Substrate
1-2
Applications
* Residential and commercial CO detectors
* Air quality controllers
* Ventilation control for indoor parking garages
1-3
2
4
1
Lead pin
Top view of the sensor
without cap
Structure
Figure 1 shows the structure of TGS2442. The sensor
utilizes a multilayer structure. A glass layer for thermal
insulation is printed between a ruthenium oxide (RuO2)
heater and an alumina substrate. A pair of Au electrodes
for the heater are formed on a thermal insulator. The gas
sensing layer, which is formed of tin dioxide (SnO2), is
printed on an electrical insulation layer which covers the
heater. A pair of Pt electrodes for measuring sensor
resistance is formed on the electrical insulator. An activated
charcoal filter is used for the purpose of reducing the
influence of noise gases.
3
Metal base
Noble metal
electrode
1-4 Basic measuring circuit
Figure 2 shows the basic measuring circuit of the TGS2442.
Circuit voltage (Vc) is applied across the sensing element
which has a resistance (Rs) between the sensor’s two
electrodes (pins No. 2 and No. 3) and a load resistor (RL)
connected in series. The sensing element is heated by the
heater which is connected to pins No. 1 and No. 4.
Figure 1 - Sensor structure
The sensor requires application of a 1 second heating cycle
which is used in connection with a circuit voltage cycle of
1 second. Each VH cycle is comprised by 4.8V being applied
to the heater for the first 14ms, followed by 0V for the
remaining 986ms. The Vc cycle consists of 0V applied for
995ms, followed by 5.0V for 5ms. For achieving optimal
sensing characteristics, the sensor's signal should be
measured after the midpoint of the 5ms Vc pulse of 5.0V
(for illustration, see the timing chart in Fig. 3).
NOTE: Application of a Vc pulse condition is required to
prevent possible migration of heater materials into the
sensing element material. Under extreme conditions of high
humidity and temperature, a constant Vc condition could
result in such migration and cause long term drift of Rs to
higher values. A 5ms Vc pulse results in significantly less
driving force for migration than a constant Vc condition,
rendering the possibility of migration negligibly small.
Revised 03/00
Figure 2 - Basic measuring circuit
(including equivalent circuit)
2
TECHNICAL INFORMATION FOR TGS2442
1-5 Circuit &␣ operating conditions
The following conditions should be maintained to
ensure stable sensor performance:
Model number
TGS 2442
Sensing element type
M1
Standard package
TO-5 metal can
Target gases
Carbon monoxide
Typical detection range
Standard circuit
conditions
Electrical characteristics
under standard test
conditions
30 ~ 1000 ppm
Heater voltage cycle
VH
VHH=4.8V±0.2V DC, 14ms
VHL=0V, 986ms
Circuit voltage cycle
VC
VC=0V for 995ms,
VC=5.0V±0.2V DC for 5ms
Load resistance
RL
variable (≥10kΩ)
Heater resistance
RH
17 ± 2.5Ω at room temp.
Heater current
IH
approx. 203mA(in case of VHH)
Heater power
consumption
PH
approx. 14mW (ave.)
Sensor resistance
Rs
6.81 kΩ ~ 68.1 kΩ in 100ppm of
carbon monoxide
Sensitivity
(change ratio of Rs)
β
Test gas conditions
Standard test conditions
Figure 3 - Circuit voltage and heater voltage cycles
0.23 ~ 0.49
Top View
Carbon monoxide in air
at 20±2˚C, 65±5%RH
Same as Std. Circuit Condition
(above)
Circuit conditions
Conditioning period before
test
> 2 days (under review)
Formula for calculation of sensor resistance:
Rs
=
ø9.2±0.2
Vc x RL - RL
Vout
ø8.1±0.2
ø0.3
Sensitivity (change ratio of Rs) is calculated with two measured values of
Rs as follows:
β = Rs (CO,300ppm)
Rs (CO,100ppm)
5.9
ø6.0+0,-0.3
12.7±0.5
To facilitate usage of this sensor, TGS2442 is shipped in presorted
groupings which have a more narrowly defined range of β:
Code:
A: 0.23~ 0.34
B: 0.26 ~ 0.37
C: 0.29 ~ 0.40
D: 0.32 ~ 0.43
E: 0.35 ~ 0.46
F: 0.38 ~ 0.49
0.2
1.5
The above six classification will be further subdivided into the
following rankings of Rs values in 100ppm of CO:
Code:
1: 6.81 ~ 21.5kΩ
2: 10.0 ~ 31.6kΩ
3: 14.7 ~ 46.6kΩ
4: 21.5 ~ 68.1kΩ
10.0±1.0
ø0.55±0.05
1-6 Mechanical Strength
The sensor shall have no abnormal findings in its
structure and shall satisfy the above electrical
specifications after the following performance tests:
Withstand Force - withstand force > 5kg in each
(pin from base) direction
Vibration - frequency--10-500Hz (equiv. to
10G), duration-6 hours, x-y-z
direction
Shock - acceleration-100G, repeat 5 times
45˚
4
1
3
2
ø5.1±0.1
Bottom view
Figure 4 - Dimensions
1-7 Dimensions (see Fig. 4)
Revised 03/00
3
TECHNICAL INFORMATION FOR TGS2442
2. Operation Principle
400
The optimum conditions of sensitivity and selectivity
to CO of the TGS2442 occurs at sensor temperatures
less than 100˚C. However, at these lower temperatures, the sensing element may be influenced by
humidity and other contaminants, so the sensing
element requires periodic heat cleaning at more than
300˚C. As a result, the TGS2442 is pulse heated to
achieve optimal sensing characteristics at low
temperatures.
350
Temperature (˚C)
300
250
200
150
100
50
0
0
5
10
VH pulse width (msec)
15
20
Figure 5 - Relationship of sensor temperature and
VH pulse width (VH=4.8V, RH=17Ω)
A signal detection point of 997.5ms into the VH pulse
cycle (ref. timing chart in Fig. 3) is used to obtain the
optimum combination of gas sensitivity performance
and minimized ambient humidity effect.
14ms VH pulse
Air
Rs/Ro
100
CO 30ppm
10
CO 100ppm
Detection point
(997.5 msec)
1
CO 300ppm
0.1
0
200
400
600
Time (msec)
800
1000
Figure 6a - Sensitivity characteristics during the VH pulse cycle
(Ro = Rs in 100ppm CO at 997.5ms of VH cycle)
1000
14ms VH pulse
Detection point
(997.5 msec.)
Temp (˚C)
Figure 6a shows the pattern of resistance change ratio
(Rs/Ro) for various CO concentrations which occurs
during the 1 second heater pulse cycle, starting with
application of the 14ms VH pulse. During the VH
pulse, initially sensor resistance drops quickly and
then returns to a higher value. After the VH pulse
concludes and the sensing element’s surface
temperature begins to decrease (Fig. 6b), sensor
resistance reaches to its maximum value and then
begins to decline. Note that the shape of the response
pattern varies according to the concentration of CO-higher CO concentrations result in a minimum Rs/
Ro value which occurs more quickly and has a lower
value. In addition, shortly after the VH pulse, the
Rs/Ro value also trends downwards at a greater rate
for higher CO concentrations after reaching its
maximum value.
100
10
0
200
400
600
Time (msec)
800
1000
Figure 6b - Surface temperature of sensing element
during VH pulse cycle
(0msec = start of VH pulse cycle)
Revised 03/00
4
TECHNICAL INFORMATION FOR TGS2442
100
3. Basic Sensitivity Characteristics
Air
Ethanol
3-1 Sensitivity to various gases
As shown by Figure 7, TGS2442 shows very good
sensitivity to CO since the sensitivity curve to CO
shows a sharp drop in sensor resistance as CO
concentration increases. In comparison, sensitivity
to ethanol (C2H5OH) is very low as evidenced by the
relatively flat slope of its sensitivity curve and high
resistance values.
H2
1
0.1
CO
0.01
1
10
100
1000
10000
Gas Concentration (ppm)
Figure 7 - Sensitivity to various gases
(Ro = Rs in 100ppm of CO)
1000
50%RH
100
Rs/Ro
The amount of CO␣ generated by cigarette smoke is
roughly equivalent to 20ppm of CO when 10
cigarettes are smoked in a room of roughly 24 cubic
meters in size. As a result, the influence of cigarette
smoke itself would not be sufficient to cause the
sensor to generate an alarm for residential detectors
using TGS2442 which are normally calibrated to
alarm at 100ppm of CO.
10
Rs/Ro
Figure 7 shows the sensor’s relative sensitivity to
various gases. The Y-axis shows the ratio of sensor
resistance in various gases (Rs) to the sensor
resistance in 100ppm of CO (Ro).
Air
10
3-2 Temperature and humidity dependency
Figure 8a shows the temperature dependency of
TGS2442. The Y-axis shows the ratio of sensor
resistance for various CO concentrations under
various temperature conditions (Rs) to the sensor
resistance in 100ppm of CO at 50%RH (Ro).
Figure 8b shows the humidity dependency of
TGS2442. The Y-axis shows the ratio of sensor
resistance for various CO concentrations under
various relative humidity conditions (Rs) to the
sensor resistance in 100ppm of CO at 20˚C.
CO 30ppm
1
CO 100ppm
CO 300ppm
0.1
-20
-10
0
10
20
30
40
50
Figure 8a - Temperature dependency at 50%RH
(Ro = Rs in 100ppm of CO at 20˚C)
100
Air
20˚C
10
Rs/Ro
An inexpensive way to compensate for temperature
dependency to a certain extent would be to
incorporate a thermistor in the detection circuit (please
refer to Section 5-2).
60
Temperature (˚C)
CO 30ppm
1
CO 100ppm
CO 300ppm
0.1
0
20
40
60
80
100
Humidity (%RH)
Figure 8b - Humidity dependency at 20˚C
(Ro = Rs in 100ppm of CO at 50%RH)
Revised 03/00
5
TECHNICAL INFORMATION FOR TGS2442
3-3 Gas response pattern
air
70ppm CO
150ppm CO
air
400ppm
CO
air
140
160
1000
Rs (kΩ)
The response pattern of the TGS2442 sensor is similar
to that of a constant VH sensor since data acquisition
is carried out once every second during operation.
Figure 9 shows the pattern of the output signal when
the sensor is placed into 70, 150, and 400ppm of CO
and then returned to normal air. The response time
to 90% of the saturated signal level is roughly 3.3
minutes and the recovery of the sensor to 90% of the
base level is within 10 minutes. This data
demonstrates that TGS2442 possesses sufficient
response speed for meeting UL requirements for CO
detectors.
100
10
0
20
40
60
80
100
Time (min)
120
Figure 9 - TGS2442 response pattern in 70, 150, and 400ppm CO
Rs/Ro
The TGS2442 should be used with a heater voltage
of 4.8 ± 0.2V. Although at the detection point the
sensor’s temperature is close to room temperature,
its gas sensing characteristics are affected by heater
voltage as can be seen in Figure 10. At lower heater
voltage, the Rs/Ro decreases and the difference in
sensitivity between CO concentrations narrows. If
heater voltages is higher, sensor resistance increases.
CO 30ppm
CO 100ppm
CO 300ppm
10
3-4 Heater voltage dependency
1
0.1
4.2
4.4
4.6
4.8
5.0
Heater voltage (V)
5.2
5.4
Figure 10 - Heater voltage dependency
(Ro = Rs at 100ppm of CO and VH=4.8V)
Revised 03/00
6
TECHNICAL INFORMATION FOR TGS2442
3-5 Initial action
10
After energizing, the sensor’s resistance reaches to
90% of its final value in less than one minute, an alarm
delay circuit should be incorporated into detectors
using TGS2442 to prevent activation of an alarm
during this period.
Rs/Ro
Figure 11 shows the initial action of the sensor’s
resistance. For purposes of this test, the sensor was
stored unenergized in normal air for 40 days after
which it was energized in clean air.
1
0.1
0
5
10
15
Time (min.)
20
25
30
Figure 11 - Initial action
(Ro = Rs after 20 minutes of energizing)
3-6 Influence of unenergized storage
This chart shows that after energizing, unlike the
sensor resistance in clean air (as shown in Fig. 11),
resistance in CO first decreases slightly and then
returns to a stable level, demonstrating the need for
adhering to the recommended preheating period
prior to calibration.
10
Rs/Ro
Figure 12 shows the influence of unenergized storage
on sensor resistance. Sensors were stored
unenergized in normal air for 30 days after which
they were energized. The Y-axis represents the ratio
of sensor resistance in various concentrations of
CO␣ after the unenergized period (Rs) to the resistance
in 100ppm of CO after energizing at the rated voltage
for 4 days (Ro).
CO 30ppm
CO 100ppm
CO 300ppm
1
0.1
0
2
4
6
8
10
12
14
Time (days)
Figure 12 - Time dependency
(Ro = Rs in 100ppm of CO after 4 days, n=20)
Revised 03/00
7
TECHNICAL INFORMATION FOR TGS2442
4. Reliability
4-1 Interference gas test
80
Rs/Ro
Tests conducted in this section demonstrate that TGS2442
can meet the requirements of various testing standards
without incurring adverse long term effects from such tests.
100
Figure 13 shows the results of testing the TGS2442
sensor for durability against various interference
gases specified by GRI Test Protocol 1. The test was
conducted by exposing the sensor to each gas shown
in Fig. 13 (starting with CO 100ppm) for two hours,
then removing the sensor to fresh air for just under
one hour, and followed by inserting the sensor into
the next test gas. This procedure was repeated for
the full range of gases shown in Fig. 13.
Figure 15 demonstrates that sensor subjected to the
interference gas tests do not change their
characteristics after exposure to these gases. Samples
subjected to the interference gas test were compared
to reference samples (not subjected to the interference
gas tests). Over a two week period, when not
undergoing gas test, all samples were energized in
fresh air under standard circuit conditions.
This data suggests that TGS2442 shows good
durability against every gas used in GRI Test Protocol
1 and meets the requirements of UL2034.
Revised 03/00
20
0
CO M Bu He Et Is CO Am Ac Et To Tr Ni Ni Et CO Ac
e
i
t
t
h o
h
h
10 tha tan pta yl a pro 2 5 mo eto an luen chlo ric rou yle 10 rylic
0p ne e 3 ne ce pa 00 ni ne ol e ro oxi s o ne 0p c
pm 50 00 50 ta no 0p a 1 20 20 20 eth de xid 20 pm em
en
0p pp 0p te l 2 pm 00 0p 0pp 0pp an 50 e 0p
pp pm m m e
t1
p 2
pm m pm 20 00
p
0p pp
20 pm 00p m
m
0g
pm m
0p
pm
pm
Fig. 13 - Effects of interference gases (GRI Test Protocol 1)
(Ro = Rs in CO 100ppm)
initial value
80
after 2 hrs.
70
60
50
Rs/Ro
Figures 14 and 15 show data from tests conducted
using interference gases listed in the UL2034 standard
for Selectivity Test (Sec. 38) which are deemed to
represent air contaminants likely to be found in the
vicinity of an installed CO detector. Data for Figure
14 was collected by exposing samples in each of the
test gases for a period of two hours as required by
the UL standard. Sensor resistance at both the initial
point and at the conclusion of test gas exposure was
recorded. When compared to the sensor’s measured
resistance in 100ppm of CO (Ro), in all cases the
resistance in test gas remained at more than 40 times
that of resistance in 100ppm of CO, showing the
sensor to have negligible influence by these gases.
40
40
30
20
10
0
Methane
500ppm
Isobutane
300ppm
Heptane
500ppm
CO2 Ethyl acetate Isopropanol
5000ppm
200ppm
200ppm
Figure 14 - Selectivity test (UL2034, Sec. 38)
(Ro = Rs in CO 100ppm)
100
CO 30ppm (ref)
CO 100ppm (ref)
CO 300ppm (ref)
CO 30ppm (test)
CO 100ppm (test)
CO 300ppm (test)
test
10
Rs/Ro
Because the sensor is exposed to each of the test gases
consecutively, to some small extent the effect of the
previous test gas may affect subsequent tests for a
short period. However, despite the short-term effects
of such gases remaining after exposure, the sensor
still shows significantly less sensitivity to each test
gas when compared to 100ppm of CO, and CO
sensitivity remains unaffected.
60
1
0.
0
2
4
6
8
10
12
14
Time (days)
Figure 15 - Stability of sensors subjected to selectivity test
(Ro = Rs in CO 100ppm on Day 6)
8
TECHNICAL INFORMATION FOR TGS2442
Figure 16 shows long-term stability data for TGS2442.
Test samples were energized in normal air and under
standard circuit conditions. Measurement for
confirming sensor characteristics was conducted
under standard test conditions (20˚C, 65%RH). The
initial value was measured after four days of
energizing in normal air at the rated voltage. The Yaxis shows the ratio between measured sensor
resistance and the initial (Day 4) resistance value,
each in 100ppm of CO. Additional data will be
published as it becomes available.
Air
CO 30ppm
CO 100ppm
CO 300ppm
H2 1000ppm
EtOH 1000ppm
1000
100
Rs/Ro
4-2 Long-term stability
10
1
0.1
0
50
100
150
200
250
300
350
Time (days)
Figure 16 - Long term stability
(Ro = Rs in 100ppm of CO at Day 4)
CO 30ppm (ref)
CO 100ppm (ref)
CO 300ppm (ref)
CO 30ppm (test)
CO 100ppm (test)
CO 300ppm (test)
4-3 Corrosion test
10
Rs/Ro
To demonstrate the durability of TGS2442 against
corrosion, samples were subjected to test conditions
called for by UL2034, Sec. 57-Corrosion Test. Over a
three week period, a mixture of 100ppb of H2S, 20ppb
of Cl2, and 200ppb of NO2 was supplied to the sensor
at a rate sufficient to achieve an air exchange of 5
times per hour. When compared to reference samples
not subjected to these corrosive gases, no significant
difference can be noticed.
Test
1
0.1
0
5
10
15
20
25
30
35
40
Time (days)
Figure 17 - Durability against corrosion
(Ro = Rs in 100ppm of CO at Day 2)
CO 30ppm (ref)
CO 100ppm (ref)
CO 300ppm (ref)
CO 30ppm (test)
CO 100ppm (test)
CO 300ppm (test)
4-4 Variable ambient temperature test
10
Rs/Ro
To show the ability of TGS2442 to withstand the
effects of high and low temperatures representative
of shipping and storage, the sensor was subjected to
the test conditions of UL2034 Sec. 44.2-Effect of
Shipping and Storage. Unenergized test samples
were subjected to 70˚C for 24 hours, allowed to cool
to room temperature for 1 hour, subjected to -40˚C
for 3 hours, and then allowed to warm up to room
temperature for 3 hours. When compared to
reference samples not subjected to these temperature
extremes, no significant difference can be noticed.
Test
3 hrs. after test
1
0.
0
5
10
15
20
25
30
Time (days)
Figure 18 - Variable ambient temperature test
(Ro = Rs in 100ppm of CO at Day 0)
Revised 03/00
9
TECHNICAL INFORMATION FOR TGS2442
4-5 Humidity test
52˚C/95%RH
10
20˚C/15%RH
Rs/Ro
Figure 19 shows the comparison of reference sensors
to those energized and exposed in an atmosphere of
52˚C and 95% RH for a period of 168 hours, returned
to normal air for 2 days, then followed by 168 hours
in 20˚C/15%RH as required by UL2034 Sec. 46A.1Humidity Test. As the test measurements taken after
the conclusion of the Humidity Test demonstrate,
sensors subjected to the test show influence by
humidity, but the sensor quickly returns to its normal
value.
CO 30ppm (ref)
CO 100ppm (ref)
CO 300ppm (ref)
CO 30ppm (test)
CO 100ppm (test)
CO 300ppm (test)
100
1
0.1
0
5
10
15
Time (days)
20
25
Figure 19 - Humidity test
(Ro = Rs in 100ppm of CO at Day 0)
4-6 Stability test
(1) False alarm test
10
end test
Rs/Ro
To show the sensor’s behavior under continuous low
level exposure to CO, samples were tested against
the procedure detailed in UL2034, Sec. 41.1(c)Stability Test. Test samples were exposed to 30ppm
continuously for a period of 30 days under standard
circuit conditions. As this data demonstrates, false
alarming does not occur as a result of continuous low
level CO exposure.
CO 30ppm (ref)
CO 100ppm (ref)
CO 300ppm (ref)
CO 30ppm (test)
CO 100ppm (test)
CO 300ppm (test)
1
start test
0.1
0
10
20
30
40
50
60
Time (days)
Figure 20 - False alarm test
(Ro = Rs in 100ppm of CO prior to test)
CO 30ppm (ref)
CO 100ppm (ref)
CO 300ppm (ref)
CO 30ppm (test)
CO 100ppm (test)
CO 300ppm (test)
100
(2) Temperature cycle test
10
Test
Rs/Ro
In accordance with UL2034, Sec. 41.1(e)-Stability Test,
test samples were exposed to ten cycles (< 1 hr. and
> 15 min.) of temperature from 0˚C and 100%RH to
49˚C and 40%RH. As the three test measurements
taken after the conclusion of the test period
demonstrate, sensors subjected to the test show
negligible influence by temperature extremes.
after 8 hrs.
1
0
1
2
3
4
5
6
7
8
Time (days)
Figure 21 - Temperature cycle test
(Ro = Rs in 100ppm of CO prior to test)
Revised 03/00
10
TECHNICAL INFORMATION FOR TGS2442
5. Circuit Examples
5-1 Basic circuit including trouble detection
Figure 22 - Basic circuit (including trouble detection)
5-2 Calibration and temperature/humidity compensation
Figure 23 - Calibration & temperature compensation
S-In
8-bit A/D input port
H-In
8-bit A/D input port
R-In
8-bit A/D input port
(Each of these ports acquires data at preset timing.)
RL1,RL2
open drain output port
Pulse
open drain output port
Temp
open drain output port
Calb
open drain output port
Circuit voltage (Vc) across the sensor is applied when RL1 or RL2
ports give the Low (L) output signal at preset timing.
V1
sampling voltage for gas detection
V2
sampling voltage for sensor element trouble detection
V3
sampling voltage for reference voltage
V4
sampling voltage for temperature compensation
V5,V6
sampling voltage for detecting heater circuit breakage
V1 and V2 are acquired during the last half of the 5msec VC pulse
(the first half of the VC pulse is considered as the transient period).
V5 is acquired during heater OFF, and V6 is acquired during heater
ON pulse.
RA = 50% of Vc at the targeted gas concentration
RB = 200-300kΩ
RC = 10kΩ
RD = 20kΩ
RE = 10kΩ
VRadj = 100kΩ
Thermistor: R (25˚C) = 10kΩ, B constant = 3400
Revised 03/00
Sensor resistance (Rs):
Rs =
Calibration resistance
(RCalb):
RCalb =
5 - V1
V1
5 - V3
V3
Coefficient for temp.
compensation (KTemp):
KTemp =
5 - V4
V4
Operation state:
State
Conditions
Normal
Rs x KTemp > RCalb and V2 ≥ 0.5V
Alarm
Rs x KTemp ≤ RCalb and V2 ≥ 0.5V
Sensor malfunction
V2 < 0.5V
Heater malfunction
V5 ≥ 0.1V or V6 ≤ 4.5V
11
Example of Application Circuit for TGS2442
TECHNICAL INFORMATION FOR TGS2442
Revised 03/00
12
TECHNICAL INFORMATION FOR TGS2442
6. Marking and Packaging
6-1 Batch number
Rank and Lot. No. are indicated on the shrink wrap
on the side of the sensor cap as shown in Figure 24.
Rank Lot No.
A1
001512
Trace no.
Week (01-52)
Gas testing year (00-99)
Based on Rs 100ppm CO (1-4)
Based on β value (A-F)
Figure 24 - Batch number coding
6-2 Packaging method (Fig. 25)
Fifty (50) pieces of sensor are packed in a plastic
container, and five (5) containers are sealed inside a
moistureproof aluminum coated bag. Several bags
are then packed in a carton box (see Fig. 26).
x5
Plastic container for 50pcs.
(188 x 188mm)
6-3 Label
A label showing product name, Rank and Lot No.,
quantity, and inspection date is affixed to the
aluminum coated bag.
6-4 Handling instructions
It is recommended to begin sensor preheating within
24 hours after opening a sealed bag. Please keep
unused sensors in a tightly sealed moistureproof bag.
Label
Aluminum coated bag - holds 250pcs.
(265 x 300mm)
Figure 25 - Packing methodology
Bag
(250 pcs.)
2000pcs. (250 pcs. x 8 bags)
Figure 26 - Carton box
Revised 03/00
13
CALIBRATION OF TGS2442
Signal Processing and
Calibration Techniques
for CO Detectors
Using TGS2442
This brochure offers users
important technical advice for
handling and calibration of the
TGS2442 CO sensor, including
calibration techniques using a
microprocessor. Detector circuit
and calibration facility design
should be carried out with these
points in mind.
Page
Basic Circuit Structure..................................................................2
Method of Signal Processing
Basic flow.............................................................................2
Detection of sensor-related trouble conditions............................2
Calculation of CO concentration
Sensor resistance (Rs) calculation..................................4
Temperature compensation of ƒ(Rs)..................................4
Compensation for variation in Rs grades......................5
Compensation for variation within β grades......................5
Converting ƒ(Rs➁) to CO concentration output...............5
Calibration Using Classified Sensors
Preheating...................................................................................6
Pre-calibration
Select a load resistor............................................6
Compensation according to β grade.................................6
Main calibration....................................................................7
Appendix-Calibration at two or more concentrations...................8
Ability of Sensor to Meet Performance Standards
Requirements of performance standard.......................................8
Estimation of calibration accuracy.................................9
Conclusion........................................................................9
IMPORTANT NOTE: OPERATING CONDITIONS IN WHICH FIGARO SENSORS ARE USED WILL VARY
WITH EACH CUSTOMER’S SPECIFIC APPLICATIONS. FIGARO STRONGLY RECOMMENDS
CONSULTING OUR TECHNICAL STAFF BEFORE DEPLOYING FIGARO SENSORS IN YOUR APPLICATION
AND, IN PARTICULAR, WHEN CUSTOMER’S TARGET GASES ARE NOT LISTED HEREIN. FIGARO
CANNOT ASSUME ANY RESPONSIBILITY FOR ANY USE OF ITS SENSORS IN A PRODUCT OR
APPLICATION FOR WHICH SENSOR HAS NOT BEEN SPECIFICALLY TESTED BY FIGARO.
Revised 04/01
1
CALIBRATION OF TGS2442
In order to meet the various regulations to which
CO␣ detectors are subject, circuit design and method
of calibration should be based on the performance of
the gas sensor. Recommended procedures for usage
of the TGS2442 CO␣ sensor are set forth in this
document.
1. Basic Circuit Structure
Circuit structure needed to satisfy performance
standards should provide for complex signal processing, necessitating usage of a microcomputer with
features such as those shown in Table 1. This
microcomputer will process multiple signals from the
sensor, a thermistor (for temperature compensation),
and a potentiometer (for calibration). The microcomputer will output a control signal for alarm operation.
The following six types of output signals should be
monitored (refer to the sample circuit diagrams and
timing charts in Figs. 1 and 2 on the facing page):
V1: Sampling voltage for gas detection
V2: Sampling voltage for sensor element trouble detection
V3: Sampling voltage for reference voltage
V4: Sampling voltage for temperature compensation
V5,V6: Sampling voltage for detecting heater trouble
CPU core
4 or 8 bit microcomputer
(with 2~8MHz clock)
Memory size
2~4k byte ROM
128~256 nibble or byte RAM
Pin size
28~32 pins
Option
8-bit A/D converter
(more than 3 channels)
high current direct drive
(more than 6 ports)
Mode
selection port
select calibration or normal
operation modes
Table 1 - Recommended microcomputer features
2. Method of Signal Processing
2-1. Basic flow
Figure 3 shows the basic flow of detector operation.
To avoid potential nuisance alarming during the
sensor’s initial action period, a warm-up period of
several minutes (alarm delay) should be utilized
upon initial powering of the detector. After this
period, the program in the microcomputer starts the
main gas detection routine. During the gas detection
routine, the above mentioned six output signals are
acquired during each one second interval. V2, V5 and
Revised 04/01
Power ON
Alarm delay for initial action
V1~V6 sampling
Generate trouble
signal
Yes
Sensor trouble
detection
No
CO calculation
Display CO concentration
*
Main routine
(1 sec. cycle)
Convert to COHb
concentration
Generate alarm
Yes
Alarm
determination
No
Suppress alarm
* if applicable
Figure 3 - Basic flow of detector operation
V6 are used for detecting sensor-related trouble and
heater driving circuit malfunction while V1, V3 and
V4 are used to calculate CO gas concentration. This
concentration is subsequently converted by timeweighting into a COHb concentration (the key
measurement used in performance standards for
determining the generation of alarm signals).
2-2. Detection of sensor-related trouble conditions
The sensor trouble mode should indicate that the
sensor’s heater has broken or that the sensor element
itself has been damaged. The trouble signal for heater
breakage can be detected by an abnormal rise in
heater resistance or as a result of lead wire breakage,
transistor problems, and/or short of the heater. The
sensor element trouble signal is generated by damage
to the sensing material or breakage of a lead wire.
These phenomena produce an extreme change to the
values of V5, V6 and V2; consequently monitoring
V5, V6 and V2 enables detection of sensor-related
trouble conditions. See Table 2 for recommended
conditions for monitoring sensor-related trouble.
Heater trouble
V5 ≥ 0.1V
V6 ≤ 4.5V
Sensor element damage
V2 ≤ 0.15V
Table 2 - Conditions under which an trouble signal
should be generated
2
CALIBRATION OF TGS2442
Figure 1 - Basic circuit (including trouble detection)
Figure 2 - Calibration & temperature compensation
Legend of Circuit Diagrams
S-In
H-In
R-In
8-bit A/D input port
8-bit A/D input port
8-bit A/D input port
(Each of these ports acquires data at preset timing.)
RL1
RL2
Pulse
Temp
Calb
open drain output port
open drain output port
open drain output port
open drain output port
open drain output port
Revised 04/01
Circuit voltage (Vc) across the sensor is applied when RL1 or RL2
ports give the Low (L) output signal at preset timing.
V1 and V2 are acquired during the last half of the 5msec VC pulse
(the first half of the VC pulse is considered as the transient period).
V5 is acquired during heater OFF, and V6 is acquired during heater
ON pulse.
RA = closest value to Rs at calibrated CO concentration
RB = 300kΩ
RC = 10kΩ
RD = 20kΩ
VRadj = 100kΩ
Thermistor: R (25˚C) = 15kΩ, B constant = 4200
3
CALIBRATION OF TGS2442
100
2-3. Calculation of CO concentration
Figure 4 illustrates the process for calculating CO
concentration from V1, V3 and V4 signals.
40%RH
10
a. Calculation of ƒ(Rs) value
b. Temperature compensation of ƒ(Rs) value
c. Compensation for Rs grade variation
d. Compensation for β grade
e. Convert to CO concentration output
Rs/Ro
Sample V1, V3, V4
1
0˚C
15˚C
25˚C
35˚C
40˚C
50˚C
0.1
0.01
100
Figure 4 - Signal processing flow for
calculation of CO concentration
1000
CO concentration (ppm)
Figure 5a - Temperature dependency of Rs
(Ro = Rs at 100ppm CO, 25˚C/40%RH)
a. Sensor resistance (Rs) calculation:
To represent sensor resistance (Rs) value in the
microcomputer, the expression ƒ(Rs) shall be used:
100
40%RH
ƒ(Rs) = ( 5 - V1 ) / V1
KTemp = Rs/Ro, where
Rs = actual sensor resistance in various conditions
Ro = actual sensor resistance under standard conditions
Using the circuit condition and recommended
thermistor in Figs. 1 & 2, measure actual Rs values
at various ambient temperatures and sample
thermistor output (V4). The correlation between V4
and KTemp can be obtained by this procedure and
should be written as a table of coefficients into ROM
on the microcomputer. Then, by monitoring V4
output during normal operations, the proper KTemp
value can be selected from the table and used to
temperature compensate ƒ(Rs) values according to
the following formula:
ƒ(Rs①)
= ƒ(Rs) / KTemp
ƒ(Rs①) represents temperature compensated ƒ(Rs).
Table 3 shows an example coefficient table for temperature compensation where measuring 100ppm of
CO in 25°C/40%RH is used as a standard condition.
Figures 5a and 5b show typical values of ƒ(Rs) before
and after temperature compensation.
Revised 04/01
Rs/Ro
ƒ(Rs①)/ƒ(Ro)
10
b. Temperature compensation of ƒ(Rs)
In order to compensate for the temperature
dependency of ƒ(Rs) value, a temperature compensation coefficient (KTemp) must be determined.
1
0.1
0~50˚C
0.01
100
1000
CO concentration (ppm)
Figure 5b - ƒ(Rs) value compensated for temperature [ƒ(Rs①)]
(Ro = Rs at 100ppm CO, 25˚C/40%RH)
V4 Input Value
Temp (˚C)
KTemp Coefficient
4.15
-10
2.72
3.92
-5
2.34
3.66
0
1.99
3.37
5
1.70
3.06
10
1.46
2.75
15
1.27
2.44
20
1.12
2.14
25
1.00
1.86
30
0.909
1.61
35
0.839
1.38
40
0.786
1.18
45
0.744
1.01
50
0.712
0.86
55
0.687
0.73
60
0.668
Table 3 - Table of temperature compensation coefficients
4
CALIBRATION OF TGS2442
ƒ(Rref) = ( 5 - V3 ) / V3
ƒ(Rs➁) is defined as a variation compensated ƒ(Rs①)
value:
10
Rs/Ro
①)/ƒ(Ro)
ƒ(Rs
c. Compensation for variation within Rs grades
Variation in absolute Rs values among sensors within
any given grade should be normalized by adjusting
ƒ(Rref) as part of the calibration process.
1
max
center
0.1
min
ƒ(Rs➁) = ƒ(Rs①) / ƒ(Rref)
To calibrate the sensor correctly, ƒ(Rs➁) should be
equal to 1.0 at the desired concentration. To do this,
change V3 values by adjusting VRadj. Figures 6a and
6b show the variation within Rs grades and the result
of compensating for such Rs variation.
0.01
10
100
1000
CO Concentration (ppm)
Figure 6a - Rs variation within Rs grades
(Ro = Rs at 100ppm CO, 25˚C/40%RH)
Rs/Ro
➁≠)/ƒ(Ro)
ƒ(Rs
Rs
/Ro
10
d. Compensation for variation within β grades
Variation in sensitivity slopes (α) among sensors
within any β grade should be compensated in the
calibration process. This requires that a second ƒ(Rs➁)
1
value (at 300ppm of CO, for example) be obtained:
* measure actual ƒ(Rs) at second CO␣ concentration
* calculate ƒ(Rs➁) for the second concentration
* using the ƒ(Rs➁) values for the two concentrations,
0.1
calculate the slope (α):
α=
log ƒ(Rs➁) (300ppm) - log ƒ(Rs➁) (100ppm)
log 300 - log 100
0.01
* store the α value in the microcomputer
e. Converting ƒ(Rs➁) to CO concentration output
Using the sensitivity curve slope (α) determined
above, actual CO␣ concentration (C) can be calculated
for usage in conversion to COHb:
10
100
1000
CO Concentration (ppm)
Figure 6b - Rs variation compensation
(Ro = Rs at 100ppm CO, 25˚C/40%RH)
10
➁)/ƒ(Ro)
≠ /Ro
ƒ(RsRs
Rs/Ro
C = 100 x ƒ(Rs➁)1/α
1
max
center
min
0.1
10
Revised 04/01
100
CO Concentration (ppm)
Figure 7 - α variation within β grades
(Ro = Rs at 100ppm CO, 25˚C/40%RH)
1000
5
CALIBRATION OF TGS2442
3. Calibration Using Classified Sensors
Rs in CO 100ppm (kΩ)
Min.
Center
Max.
Recommended
RL (kΩ)
1
6.81
12.1
21.5
12
2
10.0
17.8
31.6
18
3
14.7
26.1
46.4
27
4
21.5
38.3
68.1
39
Grade
This section describes the procedure for calibrating
at 100ppm of CO using sensors classified as shown
in Tables 4 and 5. Figure 8 illustrates the flow of the
calibration process, including the method of signal
processing which takes place in the microcomputer.
3-1. Preheating
To stabilize sensor characteristics prior to calibration,
it is recommended that sensors be preheated under
standard circuit conditions for 2 - 7 days. To shorten
preheating time and simplify this process, Figaro is
currently studying alternative methods and will issue
new recommendations shortly.
3-2. Pre-calibration (adjustment during assembly)
a. Select a load resistor (RL)
Select the load resistor (RL, same as RA in Figure 1)
based on the Rs grade shown in Table 4. For optimal
resolution, the RL used should have a value as close
as possible to the Rs grade’s center value. The sensor
output signal (V1) at the calibration concentration
should be near 2.5V which is 1/2 the value of Vc.
According to the maximum and minimum values of
Rs values for each Rs grade as shown in Table 4, V1
would fall in the following range:
1.70 v ≤ V1 (at 100ppm) ≤ 3.30 v
b. Compensation according to β grade
Each Rs grade is divided into six β grades as shown
in Table 5. Using data from this table, α values for
these grades can be determined as follows and should
be stored in the microcomputer (see Table 6):
α
Table 4 - Rs grades of TGS2442
Grade
β (Rs in 300ppm/Rs in 100ppm)
Min.
Center
Max.
A
0.230
0.285
0.340
B
0.260
0.315
0.370
C
0.290
0.345
0.400
D
0.320
0.375
0.430
E
0.350
0.405
0.460
F
0.380
0.435
0.490
Table 5 - β grades of TGS2442
β Grade
α Value
Jumper
Connection
A
-1.14
001
B
-1.05
010
C
-0.969
011
D
-0.862
100
E
-0.823
101
F
-0.758
110
Table 6 - Recommended jumper connections
for β grade compensation circuit
= log β (center) / log (300/100)
Using the α value calculated above, compensation
for β grade can be done by utilizing a circuit such as
that shown in Figure 9 and connecting the three
jumper lines as indicated in Table 6.
Figure 9 - Recommended circuit for β grade compensation
Revised 04/01
6
CALIBRATION OF TGS2442
Calibration
Preheat sensor
Pre-calibration
*
* Selection of RL - Sec. 3-2(a)
β grade compensation - Sec. 3-2(b)
Burn-in detector
Signal Processing in the Microcomputer
Set microcomputer to
Calibration Mode
Insert into gas chamber
Begin Calibration Mode
Step 1
Expose to 100ppm CO
Sensor stabilization
Step 2
Power reset ON
Sample V1 (sensor output)
and V4 (thermistor output)
Step 3
Calculate ƒ(Rs)
Determine KTemp
Calculate ƒ(Rs①) (100) and store in memory
Vent gas
Step 4
Adjust VRadj
main calibration
Continuous sampling of V3 (reference voltage)
Calculate ƒ(Rref) continuously
Adjust ƒ(Rref) to the threshold value
Deactivate Calibration Mode
on microcomputer
End calibration
Figure 8 - Calibration and signal processing in the microcomputer
3-3. Main calibration (using “Calibration Mode” in the
microcomputer )
Using the mode selection port on the microcomputer,
select “Calibration Mode”. Switch the detector
ON␣ (Reset), activating calibration mode. Signal
processing in the microcomputer during the
calibration mode proceeds as shown in Figure 8. Place
the detector in a vessel which is then filled with
100ppm of CO.
Step 1:
Stabilize sensor output V1 in 100ppm of CO for the
predetermined period stored in the microcomputer.
Step 2:
Obtain V1 and V4 at the conclusion of stabilization.
Revised 04/01
Step 3:
Calculate ƒ(Rs), KTemp, ƒ(Rs①) according to the
procedure in Section 2-3(a,b). The value f(Rs①) is
stored temporarily in the microcomputer’s RAM.
Step 4:
After ventilating the calibration gas, V3 is
continuously monitored and converted to ƒ(Rref) [see
equation in 2-3(c)]. Using a potentiometer (VRadj),
adjust ƒ(Rref) to equal ƒ(Rs①) in 100ppm of CO which
was memorized in Step 3. This procedure can be
simplified by directing the microcomputer to activate
an LED when ƒ(Rref) is set to the correct level.
For example: If ƒ(Rref) = [ƒ(Rs①)(100ppm)] ± 3%,
then “LED TURN ON” else “LED TURN OFF”
7
CALIBRATION OF TGS2442
Appendix-Calibration at two or more concentrations
Detectors which display CO concentration demand
more accurate calculation of CO concentrations. To
obtain sufficient accuracy, calibration should be
conducted at two or more CO concentrations according to the signal processing procedures in Section 2.
For illustration, an example of 300ppm and 100ppm
of CO as calibration points is used in this section.
The procedure of compensation according to β grade
in Section 3-2(b) may be omitted. Instead, ƒ(Rs①) of a
second CO concentration (300ppm in this case) is
obtained by following Steps 1~ 4 in Section 3-3. By
this procedure, the sensor’s β value can be obtained
for each detector from the following equation:
Following these procedures, each sensor can be fully
compensated for temperature and sensitivity slope.
4. Ability of Sensor to Meet Performance Standards
This section will show how TGS2442, if used in the
manner described in this document, can satisfy
CO␣ detector performance standards such as UL2034
and the CSA 6-96 standard. The following
assumptions are made for this purpose:
* no error exists in calibration gas concentration
* no adjustment error in VRadj exists
* the gas sensitivity curve is linear on a logarithmic scale between
60-600ppm of CO
* no error exists in temperature compensation
Under these conditions, accuracy of calibration will
depend upon the accuracy of ƒ(Rs) and β. When
β = ƒ(Rs①)(300ppm) / ƒ(Rs①)(100ppm)
calibration is done using two or more CO
In turn, this β value can be converted into an α value concentrations (Section 3-Appendix), calibration
by using the equation shown in Section 3-2(b). This α without error can be done at the targeted gas
concentration. However, if calibration is done only
value (αmem) is temporarily stored in the RAM␣ of the
at one gas concentration and is based on one of the
microcomputer (see Sec. 2-3(d)).
combinations of Rs and β grades supplied for the
To compensate for the slope of the sensitivity curve, sensor, since β falls within a range for each grade,
a slope compensation circuit shown in Figure 10a can verification that such variation in β falls within
be used. Since the range of values for V7 is from 0 ~ acceptable limits is required.
5V and the range of α values is from -1.4 ~ -0.6, a
4-1 Requirements of performance standard
linear relationship between the V7 and α can be
established (see Figure 10b). Then, in order to The most stringent performance standard (CSA 6compensate for α value, the V7 port is used--adjust 96) requires that an alarm be generated at not less
than 5% COHb but less than 10%␣ COHb. The
the potentiometer (VRadj2) so that the αcalb value is
specified CO␣ concentration and accumulation times
set as close to αmem as possible. This procedure can
be simplified by directing the microcomputer to
10% COHb
activate an LED when α is set to the correct level:
103.0
Center (7.07% COHb)
Alarm delay time (min.)
For example: If αcalb = αmem ± 3%,
then “LED TURN ON” else “LED TURN OFF”
-0.6
αcalb
αcalb = .16(V7)-1.4
-1.0
VRadj2
53.6
C(min)
C(max)
5% COHb
23.2
14.9
10.9
7.14
-1.4
0
1
2
3
4
5
V7 (V)
Figure 10a - Slope
compensation circuit
Revised 04/01
Figure 10b - Relationship of slope (a)
to V7
65
100
200 300 400 600
CO Concentration (ppm)
Figure 11 - Alarm times permitted by performance standard
8
CALIBRATION OF TGS2442
for each concentration are spelled out by the
standard. While the standard lists specific delay times
allowed for each CO␣ concentration, to facilitate this
discussion the permissible range of CO
concentrations at each specific delay time will be
used.
performance standards even if calibrated with only
one CO␣ concentration. This verification is based on
the aforementioned assumptions. Any factors which
may influence these assumptions should be taken
into consideration when planning actual detector
design and when designing the calibration process.
The maximum and minimum values of alarming time
permitted by the standard for each CO concentration
are plotted on the chart shown in Figure 11. The log
center between these two curves (7.07% COHb) is
also plotted. From these curves, the values of concentration width allowed at each delay time can be
determined—see Table 7 for a listing of these values.
4-3 Conclusion
As Table 8 illustrates, properly calibrated TGS2442
sensors can satisfy the requirements of current
Test Point
β Grade
C (min)
C (center)
C (max)
t65 = 103
48
65
95
t100 = 53.6
75
100
145
t200 = 23.2
145
200
270
t300 = 14.9
210
300
400
t400 = 10.9
260
400
520
t600 = 7.14
400
600
800
CO Concentration (ppm)
Table 7 - CO concentration ranges vs. alarm times
as specified by performance standard
10
C(min)
➁
ƒ(Rs
)/ƒ(Ro)
Rs/Ro
4-2 Estimation of calibration accuracy
The variation in each β grade can be illustrated as in
Figure 12. The center line is based on the β center
value which is given to each sensor grade. At the
calibration point of 100ppm, the ƒ(Rs➁) value of each
sensor would be the same (ƒ(Rs➁) = 1.0). But at 60
and 600ppm, the ƒ(Rs➁) could vary as shown in the
chart. At each concentration of CO, the center ƒ(Rs➁)
can be identified and the +/- range of ƒ(Rs➁) values
can also be determined. If the maximum and
minimum ƒ(Rs➁) values fall within the allowable
ranges of the standard (Table 7), then sensors
calibrated at one CO␣ concentration could satisfy the
requirements of the performance standards. Table 8
shows maximum and minimum CO␣ concentrations
for each sensor grade compared to the range of CO
concentrations permitted by performance standards.
Alarm time
(tconc.) in
minutes
C(max)
calibration point
1
max
center
min
0.1
10
100
CO Concentration (ppm)
1000
Figure 12 - Variation of alarm concentrations in β grades
(Ro = Rs at 100ppm CO, 25˚C/40%RH)
CO Concentration Range (ppm)
65ppm
100ppm
200ppm
300ppm
400ppm
A
60.6-69.0
100
181-224
257-358
329-500
B
60.6-69.0
100
182-224
257-359
330-501
C
60.6-69.0
100
182-224
258-359
330-502
D
60.5-69.0
100
182-225
257-360
330-504
E
60.4-69.1
100
181-225
257-362
329-507
F
60.3-69.1
100
181-225
256-365
329-511
Standard
Requirement
48 ~ 95
75 ~ 145
145 ~ 270
210 ~ 400
260 ~ 520
Table 8 - TGS2442 alarm concentrations vs. performance standard requirements
Revised 04/01
9
CALIBRATION OF TGS2442
Appendix
TGS2442 Temperature Compensation Factors
supplementary data
V4 Input Value
4.15
4.11
4.06
4.02
3.97
3.92
3.87
3.82
3.77
3.71
3.66
3.60
3.54
3.49
3.43
3.37
3.31
3.25
3.19
3.13
3.06
3.00
2.94
2.88
2.81
2.75
2.69
2.63
2.56
2.50
2.44
2.38
2.32
2.26
2.20
2.14
Temp (˚C)
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
KTemp Coefficient
2.72
2.64
2.56
2.49
2.41
2.34
2.26
2.19
2.12
2.06
1.99
1.93
1.87
1.81
1.76
1.70
1.65
1.60
1.55
1.51
1.46
1.42
1.38
1.34
1.30
1.27
1.24
1.20
1.17
1.14
1.12
1.09
1.07
1.04
1.02
1.00
V4 Input Value
2.09
2.03
1.97
1.92
1.86
1.81
1.76
1.71
1.66
1.61
1.56
1.52
1.47
1.43
1.38
1.34
1.30
1.26
1.22
1.18
1.14
1.11
1.07
1.04
1.01
0.975
0.944
0.914
0.885
0.857
0.830
0.803
0.778
0.753
0.729
Temp (˚C)
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KTemp Coefficient
0.980
0.961
0.943
0.926
0.909
0.894
0.879
0.865
0.852
0.839
0.827
0.816
0.805
0.795
0.786
0.776
0.768
0.759
0.752
0.744
0.737
0.730
0.724
0.718
0.712
0.706
0.701
0.696
0.692
0.687
0.683
0.679
0.675
0.671
0.668
KTemp
10
1
0.1
-10
Revised 04/01
0
10
20
30
40
Temperature (˚C)
50
60
10