White Paper
Building a Better Ammonia Sensor
A new and improved approach to
ammonia gas detection
Building a Better
Ammonia Sensor
One of the most common industrial chemicals, ammonia (NH3) is widely used as a
refrigeration gas for cold storage facilities, blast freezers, and other areas in food and
beverage plants. As a refrigerant, ammonia offers several advantages. The gas is abundant
and extremely efficient in refrigeration systems, requiring less energy per BTU. In addition,
its low infrared (IR) absorption profile translates into zero global warming potential.
As a result, ammonia is the preferred, low-cost, environmentally friendly refrigerant for
industrial food processes, cold storage and pharmaceutical applications.
However, because ammonia is highly corrosive, it poses a serious health risk. In the U.S.,
for example, exposure limits from the Occupational Safety and Health Administration range
from 25 to 50 parts per million (ppm). In addition, ammonia gas leaks create significant
operational risk through their potential for explosions, fire and food spoilage.
Ammonia’s low IR absorption profile makes it difficult for conventional, low-cost IR
detectors to accurately read and detect low-ppm gas levels. In addition, sensors often
face significant challenges in cold storage and food processing applications.
Ammonia sensors fall into two broad categories: solid-state and electrochemical.
Solid-state sensors are prone to false alarms in difficult conditions. Electrochemical
sensors are the most effective way to monitor low-level NH3.
Unfortunately, current electrochemical sensors rely on liquid electrolytes that can rapidly
evaporate. Some standard electrolytes are aqueous and, therefore, evaporate very
quickly in extreme environments. Others are petroleum-based and, even though they
last longer, can be continually stressed in refrigerated environments by sharp changes
in temperature and humidity.
Challenging Applications
“
Challenging Applications
The EC-FX
is designed
specifically for
the engine rooms,
processing areas,
and blast freezers
that make up
the challenging
applications
found in most
cold storage
facilities.
The most common challenges in ammonia sensing include:
• Refrigerated storage and blast freezers. In conventional freezers, chillers,
spiral freezers and other cold-storage areas, electrolytes must contend with extremely
low temperatures along with changes in humidity, due to doors opening and closing,
along with periodic, high-pressure washings. In blast freezers, similarly, sensors face
rapid reductions in temperature, along with changes in humidity during hot-water
washdowns, which can reduce the life of some sensors and challenge their accuracy.
In addition, as the electrolyte dissipates, the sensor’s gain — or level of sensitivity —
must be increased. However, the elevated gain can also increase the propensity for
false alarms, particularly during sudden humidity changes.
• Engine rooms. These areas — usually hot, dry and with high levels of background
ammonia — present a different challenge. Engine rooms can quickly deplete the
electrolytes in standard ammonia sensors, shortening their lifespans and impacting
their accuracy.
Next-Generation Ammonia Detection
Responding to the need for a resilient, long-lasting ammonia sensor, Honeywell Analytics
has engineered a new sensor — called the EC-FX — for use in our new EC-FX-NH3
transmitter, which is an evolution of our Manning EC-F9-NH3. This technological
breakthrough uses a proprietary, non-evaporative electrolyte to offer numerous
advantages over the standard formulation:
• Longer lifespan and lower costs. The sensor’s thicker, higher-viscosity
electrolyte lasts two to three years in engine rooms and up to four years in
refrigerated areas. That’s up to 18 months longer than most other ammonia sensors.
• Responsiveness. The new sensor reacts quickly to ammonia gas in both hot and
cold environments, without false alarms.
• Accuracy and stability. The sensor maintains sensitivity, accuracy and a
more consistent linear response — even after exposure to NH3 gas and extreme
fluctuations in temperature and humidity.
“
Rigorous Testing
“
Rigorous Testing
The above findings are based on a series of tests that compared the new, non-evaporative
electrolyte to the industry standard. In October 2013, Honeywell Analytics tested both
The EC-FX
should define
quality and
reliability for
the industrial
refrigeration
industry.
electrolytes in coordination with the Ammonia Safety & Training Institute at Fort Ord,
in Northern California. We compared the performance of EC-F9 transmitters featuring
the new electrolyte with the performance of gas detectors containing sensors that are
traditionally used to monitor ammonia gas.
We spanned two of each sensor type to 100 ppm (NH3 mixed with air) and one of each
to 250 ppm. We then placed both types on the floor and on a shelf in a closed concrete
room (30 by 30 feet) and exposed the sensors to approximately 15 pounds of NH3,
raising gas concentrations above 70,000 ppm.
During the test, the standard sensors degraded, and some of the detectors stopped
reporting gas, having used up their entire electrolyte reservoir. By contrast, all the new EC-FX
sensors, with the non-evaporative electrolyte, continued to accurately report ammonia.
Gas exposure reduced sensitivity in the standard sensors, significantly increasing their
gain. However, the sensors with the non-evaporative electrolyte showed only modest
increases in gain, making them less susceptible to false alarms.
After the test, when the sensors were brought back to the lab, the electrolytes in the
“
standard sensors had almost completely dissipated, making them difficult to calibrate.
Figure 1
EC-FX vs. Standard Sensor Gain Trends
EC-FX Sensor 250ppm #1 Gain
EC-FX Sensor #2 Gain
EC-FX Sensor #3 Gain
EC-FX #4 Gain
EC-FX #5 Gain
EC-FX Sensor #6 Gain
Standard Sensor 250ppm #1 Gain
Standard Sensor #2 Gain
Standard Sensor #3 Gain
328,900
264,000
GAIN
199,100
134,200
69,300
4,400
1/12
1/27
2/3
2/10
2/24
Gassing Interval
3/3
EC-FX vs. Standard Sensor nA/ppm Trend
3/13
3/21
Engine Room Performance
However, the reformulated sensors continued to function well, with only minimal
degradation. In the real world, the standard sensors would need to be replaced, while
the EC-FX sensors would simply have to be recalibrated and returned to the transmitter.
Subsequent tests showed conclusively that the electrolyte in the standard sensors was
consumed
and the electrodes
when exposed to large amounts of ammonia.
EC-FX
vs. Standard
Sensor oxidized
Gain Trends
However, in the non-evaporative sensors, the electrolyte handled more iterative reactions to
EC-FX Sensor 250ppm #1
Gain free electrons.
EC-FX
Sensor
#2signal
Gain alleviated theEC-FX
Sensor
#3 Gain
generate
This
steady
need to
increase
gain. See Figure 1.
EC-FX #4 Gain
EC-FX #5 Gain
EC-FX Sensor #6 Gain
Standard Sensor 250ppm #1 Gain
Standard Sensor #2 Gain
Standard Sensor #3 Gain
Engine Room Performance
328,900
To predict performance in engine and mechanical rooms, Honeywell Analytics
264,000
conducted laboratory tests that exposed both sensor types to 20 to 30 ppm ammonia
gas, at 140 degrees and 2 percent humidity for 11 weeks — conditions far more
GAIN
199,100
extreme than the typical. One of each sensor type was spanned to 250 ppm (NH3
mixed with air) and the rest to 100 ppm.
134,200
While the standard sensors lost virtually all sensitivity, the non-evaporative sensors
retained 95 percent of their capacity. Their output only declined, on average, from 76.5
69,300
to 70 nanoamps per ppm. Because the standard sensors lost so much sensitivity, their
gain was boosted to the maximum. In addition to being at maximum gain, the standard
4,400
1/12
sensors could no longer span to 20mA. However, because the reformulated sensors
2/3
2/10
2/24
3/3
3/13
3/21
kept their sensitivity
to ammonia,
Gassing
Interval they retained significant gain headroom. See Figure 2.
1/27
Figure 2
EC-FX vs. Standard Sensor nA/ppm Trend
EC-FX Sensor 250ppm #1 nA/ppm
EC-FX Sensor #2 nA/ppm
EC-FX Sensor #3 nA/ppm
EC-FX Sensor #4 nA/ppm
EC-FX Sensor #5 nA/ppm
EC-FX Sensor #6 nA/ppm
Standard Sensor 250ppm #1 nA/ppm
Standard Sensor #2 nA/ppm
Standard Sensor #3 nA/ppm
80.00
25.00
20.00
60.00
50.00
15.00
40.00
10.00
30.00
20.00
5.00
10.00
0.00
1/12
0.00
1/27
2/3
2/10
2/24
Gassing Interval
3/3
3/13
3/21
Standard nA/ppm
EC-FX nA/ppm
70.00
Blast Freezer Performance
We also evaluated the new electrolyte’s viscosity and electrical integrity in EC-F9 bias
bump tests. During the bump tests, we applied a brief pulse to the sensing electrode to
reduce free electrons. A healthier electrolyte will allow the sensor to bounce back faster
from this imbalance.
In standard sensors, the electrolyte dries up as it ages or is exposed to hot, dry
conditions, which impairs the sensor’s response to NH3. By comparison, the nonevaporative sensor retained both its electrolyte reservoir and its capacity to accurately
detect ammonia gas. The Honeywell Analytics non-evaporative electrolyte showed
almost no change in capacity during a bump test. Overall, the sensors with the
reformulated electrolytes lost about 5 percent of their capacity, while the standard
sensors lost about 99 percent. See Figure 3.
Blast Freezer Performance
Honeywell Analytics also investigated the sensors’ performance when faced with
sudden, sharp humidity changes. Despite massive fluctuations between 5 percent and
99 percent relative humidity (RH) — comparable to those caused by freezer washdowns
— the non-evaporative electrolyte performed exceptionally well, recovering significantly
faster than the standard electrolyte. Recovery times are critically important, as lengthy
down drifts can cause false alarms.
In “ms” for Standard Sensors
EC-FX Sensor #1
EC-FX Sensor #2
EC-FX Sensor #3
EC-FX Sensor #4
Standard Sensor #1
EC-FX Sensor #5
Standard Sensor #2
EC-FX Sensor #6
Standard Sensor #3
140
600
120
500
100
400
80
300
60
200
40
100
20
0
1/5
0
1/10
1/27
2/3
2/10
2/18
Data Intervals
2/24
3/3
3/12
3/21
In “ms” for EC-FX Sensors
EC-FX vs. Standard Sensor Comparative Bump Test
Figure 3
Figure 4: Typical RH Transient Response Curves
These graphs show the typical responses for a standard sensor versus the EC-FX sens
Blast Freezer Performance
Standard Sensor Transient Hum
(Low RH 5%) (High RH 99%) 1
0.80
0.70
mA t=0
0.60
The graphs below show the typical responses for a standard sensor versus the
EC-FX sensor at a 100ppm span gain setting.
Figure 4: Typical RH Transient Response Curves
Figure 4:Figure
Typical4:RH
Typical
Transient
RH Transient
Response
Response
Curves Curves
These graphs
Theseshow
graphs
the show
typicalthe
responses
typical responses
for a standard
for a standard
sensor versus
sensortheversus
EC-FX
thesensor
EC-FXatsensor
a 100ppm
at a span
100ppm
gainspan
setting.
gain setting.
0.80
0.80
0.70
0.70
0.70
0.70
0.60
0.60
0.60
0.60
0.50
0.50
0.50
0.50
0.40
0.40
0.30
0.30
0.20
0.20
0.10
0.00
0.00
t end t end
t up ....tWhen
up .... When
mA outmA
= mA
out t=0
= mA t=0
0.40
0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0.00
InitialInitial
mA output
mA output
t=0
t=0
0.30
5% RH
5%
toRH
99%
to RH
99% RH
0.20
0.10
0.00
99% RH
99%
toRH
5%toRH5% RH
Seconds
0
13
25
0
38
13
50
25
62
38
75
50
87
62
100
75
112
87
124
100
137
112
149
124
162
137
174
149
186
162
199
174
211
186
224
199
236
211
248
224
261
236
273
248
286
261
298
273
310
286
323
298
335
310
348
323
360
335
372
348
385
360
397
372
410
385
397
410
0.10
mA/10
mA/10
mA t=0mA t=0
0.40
2
15
28
40
53
65
78
91
103
116
128
141
154
166
179
191
204
217
229
242
254
267
0.80
mA/10
0.80
mA/10
0.50
EC-FX
EC-FX
Sensor
Sensor
Transient
Transient
Humidity
Humidity
Response
Response
(Low
RH 5%)
(High
RH 99%)
100ppm
span
(Low
RH 5%)
(High
RH 99%)
100ppm
span
2
15
28
2
40
15
53
28
65
40
78
53
91
65
103
78
116
91
128
103
141
116
154
128
166
141
179
154
191
166
204
179
217
191
229
204
242
217
254
229
267
242
280
254
292
267
305
280
317
292
330
305
343
317
355
330
368
343
380
355
393
368
406
380
418
393
406
418
mA/10
Standard
Standard
Sensor
Sensor
Transient
Transient
Humidity
Humidity
Response
Response
(Low
RH 5%)
(High
RH 99%)
100ppm
span
(Low
RH 5%)
(High
RH 99%)
100ppm
span
Seconds
Seconds
Seconds
Seconds
Compared with standard sensors,
Definitions:
Definitions:
Testing of
Testing
the Transient
of the Transient
HumidityHumidity
Response
Response
for Each for
Sensor
Each Sensor
+
t up
the Honeywell Analytics nont up
=0 =0
+
+ RH
RH Peak:
thePeak:
greatest
the null
greatest
or negative
null or negative
down-drift
down-drift
from the from
initialthe
5%initial
RH resting
5% RHpoint.
resting point.
-
- RH when
RH Energy:
Energy:
thewhen
RH switches
the RH switches
()()
from 99%from
back99%
to 5%.
backMathematically:
to 5%. Mathematically:
-
evaporative electrolyte recovers
RH Energy:
+ RH when
Energy:
thewhen
RH switches
the RH switches () ()
from 5% from
to 99%.
5% Mathematically:
to 99%. Mathematically: t end
much more rapidly when RH
instantaneously changes from
t end
5 to 99 percent. The sensor is
Definitions:
Testing
the....Transient
t up ....oftWhen
up
When Humidity
Response
for
Each
Sensor
mA outmA
= mA
out
t=0
= mA
t=0
⨜
+ ⨜
RH Energy:
when the RH switches
from 5% to 99%. Mathematically:
humidity instantaneously changes
from 99 to 5 percent.
To pinpoint sensor recovery times
RH Peak: the greatest null or negative down-drift
from the initial 5% RH resting point.
-
RH Energy: when the RH switches
from 99% back to 5%. Mathematically:
-
RH Peak: the greatest signal peak or positive up-drift
from the 99% RH resting point.
amid sudden RH changes, we
Definitions:
Definitions:
Testing of
Testing
the Transient
of the Transient
HumidityHumidity
Response
Response
for Each for
Sensor
Each Sensor
t up
()
=0
+
equally responsive when the
RH Peak:
thePeak:
greatest
the signal
greatest
peak
signal
or positive
peak orup-drift
positive up-drift
- RH
from the from
99% the
RH resting
99% RHpoint.
resting point.
t up
t end
()
t
calculated
the amount of energy
up
()()
+
+ RH when
RH Energy:
Energy:
thewhen
RH switches
the RH switches
from 5% from
to 99%.
5% Mathematically:
to 99%. Mathematically:
+
+ RH
RH Peak:
thePeak:
greatest
the null
greatest
or negative
null or negative
down-drift
down-drift
from the from
initialthe
5%initial
RH resting
5% RHpoint.
resting point.
-
- RH when
RH Energy:
Energy:
thewhen
RH switches
the RH switches
from 99%from
back99%
to 5%.
backMathematically:
to 5%. Mathematically:
()()
-
- RH
RH Peak:
thePeak:
greatest
the signal
greatest
peak
signal
or positive
peak orup-drift
positivefrom
up-drift
the from
99% the
RH resting
99% RHpoint.
resting point.
=0 =0
associated with negative down-drift and positive up-drift. As shown in Figure 4,
t end
the
standard electrolyte exhibited significant down-drift when transitioning from low
t
end
to high humidity, followed by significant up-drift when conditions were reversed.
The non-evaporative electrolyte showed only minimal deviation, making it less likely
Definitions: Testing of the Transient Humidity Response for Each S
to give a false alarm under these extreme conditions.
+ weakens,
RH Energy:they
whenrequire
the RH switches
from gain
5% to to
99%. Mathematical
As standard sensors age and the electrolyte
increased
maintain sensitivity. The RH transient response is directly proportional to these gain
increases. Since the EC-FX sensor uses a non-evaporative
electrolyte,
sensitivity
+ RH Peak: the greatest
null or negative
down-drift from the initial 5
EC-FX-NH3
Detectors
degradation is minimal and, therefore, gain increases are also reduced.
-
Energy: when the RH switches from 99% back to 5%. Mathem
As a result of its robust electrolyte, the EC-FX RH
sensor
shows minimal response to
RH transients over time. Standard sensors, on the other hand, show dramatic increases
- RH
Peak:
the greatestinsignal
peak or positive
up-drift from the 99%
in reactivity, which can cause false alarms and
fault
conditions
a detection
system.
See Figure 5.
Safety, Reliability
and Reduced Costs
New EC-FXNew
Sensor
100ppm
EC-FX
Sensor 100ppm
Aged EC-FX
Sensor
100ppm
Aged
EC-FX
Sensor 100ppm
Figure
5
0.7
0.7
0.7
+ RH Energy
RH Energy
21.78
+ RH Energy
+ RH Energy
21.78
24.73
0.6 for New 0.6
Withstanding
Fluctuations
in
Humidity
and 24.73
ageing
sensors.
0.6
+ RH Null
RH Null
0.80
0.90
+ RH Null
+ RH Null
0.80
0.90
0
0
RH Energy
0.7
0.3
RH Null
0.6
0.2
RH Energy
0.5
RH0.1
Peak
+
93.93
+7.2
–
37.47
–
7.26
93.93
7.2
37.47
7.26
0.1
0
0
310
330
349
369
388
407
427
At Honeywell Analytics, we recognize this need —
enhances longevity and reliability. This sensor,
engineered to Honeywell’s highest standards,
394
371
325
302
Seconds
255
394
278
Seconds
builds on the longstanding technological
excellence of the Manning Systems product line.
394
371
348
325
255
394
278
116
0
139
93
70
46
0
In summary,
rigorous tests have shown the reformulated electrolyte maintains sensitivity
23
0
0.2
0.1
0
19
39
58
78
97
116
136
0
155
19
175
39
194
58
213
78
233
97
252
116
272
136
291
155
310
175
330
194
349
213
369
233
388
252
407
272
427
291
310
330
349
369
388
407
427
refrigerators and blast freezers, ammonia sensors
348
mA/10
0.3
0.2
Seconds
fluctuating temperature and humidity levels in
0.4
0
0.1
Seconds
RH Energy
RH Null
RH Energy
RH Peak
209
348
232
371
0.4
0
mA/10
0.5
0.1
93
mA/10
93.93
7.2
37.47
7.26
162
302
186
325
Seconds
93.93
+ RH Energy
+7.2 RH Null
37.47
– RH Energy
7.26
– RH Peak
116
0
139
Seconds
+
+
–
–
0.6
0.2
0.3
RH Energy
0.7
RH Null
0.6
RH Energy
RH0.5
Peak
209
348
232
371
0.2
0.8
0.4
0.7
0.3
310
330
349
369
388
407
427
mA/10
0
19
39
58
78
97 mA/10
116
136
0
155
19
175
39
194
58
213
78
233
97
252
116
272
136
291
155
310
175
330
194
349
213
369
233
388
252
407
272
427
291
0.4
0
0.3
0.8
70
67.05
7.2
20.04
7.02
RH Energy
RH Null
RH Energy
RH Peak
46
+
67.05
+7.2
–
20.04
–
7.02
0.8
0.4
0.5
0
RH Energy
0.7
0.3
RH Null
0.6
0.2
RH Energy
0.5
0.1
RH Peak
to the hot, dry conditions in engine rooms or the
and we responded with a breakthrough sensor,
Aged Standard
AgedSensor
Standard
100ppm
Sensor 100ppm
featuring a non-evaporative electrolyte that
0.8
0.4
0.6
23
+
+
–
–
+
+
–
–
0.7
New Standard
NewSensor
Standard
100ppm
Sensor 100ppm
and food processing. Whether they’re exposed
must combine accuracy with long life.
Aged Standard
100ppm
AgedSensor
Standard
Sensor 100ppm
0.8
mA/10
67.05
7.2
20.04
7.02
in the challenging environments of cold storage
Seconds
162
302
186
325
0
21
42
63
84
105
126
0
147
21
168
42
189
63
210
84
231
105
252
126
273
147
294
168
315
189
336
210
357
231
378
252
399
273
294
315
336
357
378
399
New Standard
100ppm
NewSensor
Standard
Sensor 100ppm
67.05
+ RH Energy
+7.2 RH Null
20.04
– RH Energy
7.02
– RH Peak
Seconds
Seconds
Seconds
RH Energy
0.7
RH Null
0.6
RH Energy
0.5
RH Peak
Ammonia detectors perform an essential role
310
330
349
369
388
407
427
0
+
+
–
–
24.73
0.90
9.68
6.24
0
19
39
58
78
97
116
136
0
155
19
175
39
194
58
213
78
233
97
252
116
272
136
291
155
310
175
330
194
349
213
369
233
388
252
407
272
427
291
0.1
0.1
0.8
Seconds
0.2
0.1
Seconds
Safety, Reliability
Reduced Costs
9.68
302
0.2
116
255
139
278
Seconds
RH Energy
RH Null
RH Energy
RH Peak
116
255
139
278
Seconds
+
24.73
+
0.90
–
9.68
–
6.24
70
209
93
232
0.2
0.7
0.3
RH Energy
0.6
RH Null
0.2
0.5
RH Energy
0.1
RH Peak
0.4
0
0.3
+
+
–
–
70
209
93
232
0
0.3
0.4
0.7
0.3
0.6
0.2
0.5
0.1
0.4
0
0.3
23
162
46
186
21.78
0.80
9.16
6.1
9.68
– RH Energy
0.5
RH Energy
Aged
EC-FX
Aged
EC-FX
100ppm
– RH Peak
6.24
– Sensor
RH Peak
6.24Sensor 100ppm
and
0.4
23
162
46
186
RH Energy
RH Null
RH Energy
RH Peak
0
21
42
63
84
mA/10
105
126
0
147
21
168
42
189
63
210
84
231
105
252
126
273
147
294
168
315
189
336
210
357
231
378
252
399
273
+
21.78
+
0.80
–
9.16
–6.1
–
0.5
mA/10
0.7
RH Energy
0.3
0.6
RH Null0.2
0.5
RH Energy
0.1
RH Peak0.4
294
315
336
357
378
399
mA/10
+
+
–
–
9.16
mA/10
9.16
– RH Energy
0
19
39
58
78
97mA/10
116
136
0
155
19
175
39
194
58
213
78
233
97
252
116
272
136
291
155
310
175
330
194
349
213
369
233
388
252
407
272
427
291
0.5
RH Energy
New
EC-FX
New
Sensor
100ppm
RH Peak
–6.1
RH PeakEC-FX
6.1 Sensor 100ppm
0.4
mA/10
+
+
–
–
Seconds
in conditionsSeconds
even harsher
than those found in engine rooms, cold storage areas and
blast freezers. And while other sensors quickly lose sensitivity after gas exposure, the
new sensor bounces back from exposure and resumes accurate detection. Moreover,
extended longevity means fewer sensor replacements, which translates to significant
cost savings over time.
For more information about Honeywell Analytics’ new EC-FX ammonia sensor, please
Find out more
www.honeywellanalytics.com
contact Honeywell Analytics. Call 800.444.9935.
Contact Honeywell Analytics:
Americas
Honeywell Analytics, Inc.
405 Barclay Blvd.
Lincolnshire, IL 60069
USA
Toll-free: 1.800.444.9935
Fax: +1.888.967.9938
[email protected]
Technical Services
Tel: 1.800.538.0363
[email protected]
Canada
2840 2nd Ave SE
Calgary, Alberta, Canada
T2A 7X9
3580 Rue Isabelle Unit 100
Brossard, Quebec, Canada
J4Y 2R3
Toll-free: 1.800.563.2967 (select lang. + press 1)
Tel: 847.955.8200
Fax: +1.450.619.2448
[email protected]
www.honeywell.com
While every effort has been made to ensure accuracy in this publication, no responsibility can be accepted for errors or omissions.
Data may change, as well as legislation, and you are strongly advised to obtain copies of the most recently issued regulations,
standards, and guidelines. This publication is not intended to form the basis of a contract.
© 2014 Honeywell International, Inc.
Canada Technical Services
Tel: 1.800.538.0363
[email protected]
www.honeywell.com