Par 3-1 Page 7 CHAPTER 3. PERFORMANCE AND TESTING SPECIFICATIONS FOR LWES
3-1. GENERAL. This chapter contains the performance standards and testing specifications for LWES.
3-2. DEFINITIONS.
a. Root Mean Square Error (RMSE). RMSE is determined by comparing the output value with the true value of a parameter according to the following equation:
(RMSE is in the same units as the measured and true values).
b. Time Constant. After a step change in the value of a parameter measured by a sensor, the time constant is the length of time it takes the sensor to register a given percentage (63 percent unless otherwise specified) of the change.
c. Resolution. The resolution of a sensor is the value of the least significant digit that is given as sensor output.
d. Variance. For the purposes of this AC, variance is defined as the difference between the value of the reference sensor and the sensor under test.
3-3. GENERAL POWER STANDARDS. This paragraph addresses aspects of power that are applicable to the LWES as a whole.
a. Input Power. LWES equipment should operate from a 120/240 volts ( V) (±10 percent), 60 Hertz ( Hz) ac (±5 Hz), 3-wire service.
b. Loss of Power. The LWES system should return to normal operation without human intervention after a power outage. When power is restored, the system should not output erroneous data.
3-4. OPERATING ENVIRONMENT. All LWES equipment and sensors should demonstrate that they meet all operating tolerances under the operational environmental conditions this AC describes.
a. Temperature. From -40oC to +3oC (-40oF to +37.4oF).
b. Wind. Up to 85 knots (kts).
Par 3-5 Page 8
3-5. MECHANICAL WIND SPEED AND DIRECTION SENSOR (If Included in System).
a. Performance Standard.
(1) Wind Speed Sensor.
(a) Range. The sensor should respond to a threshold of 2 kts and a maximum of at least 85 kts.
(b) Accuracy. The wind speed sensor should provide an accuracy of ± 2 kts up to 40 kts. Above 40 kts, RMSE should be within ± 5 percent.
(c) Resolution. The resolution should be 1 kt.
(d) Distance Constant. The distance constant should be less than 10 meters.
(e) Threshold. 2 kts.
(2) Wind Direction Sensor. This sensor should be aligned to true north.
(a) Range. 1° to 360° in azimuth.
(b) Threshold. 2 kts.
(c) Accuracy. Within ± 5° (RMSE).
(d) Resolution. To nearest 1°; dead band not to exceed 7° wide.
(e) Time Constant. Less than 2 seconds.
b. Performance Testing. If the manufacturer of the sensor has performed testing and it has been shown to meet the standards stated above, it is sufficient to submit documentation from the manufacturer establishing these facts, and further testing is not necessary. If testing is necessary, testing should be conducted in a calibrated wind tunnel, except for the wind direction accuracy test that is conducted on a bench test fixture. The wind speed sensor should be compared against a calibrated pitot-static tube or transfer reference standard traceable to the National Institute of Standards and Technology (NIST).
3-6. ULTRASONIC WIND SENSOR (If Included in System).
a. Performance Standard.
(1) Wind Speed.
(a) Range. The sensor should respond to a threshold of 1 kt and a maximum of at least 85 kts.
(b) Accuracy. The wind speed sensor should provide an accuracy of ± 1 kt up to 40 kts. Above 40 kts, RMSE should be within ± 3 percent.
Par 3-6 Page 9
(c) Resolution. The resolution should be 1 kt.
(d) Distance Constant. The distance constant should be less than 1 meter.
(e) Threshold. 1 kt.
(2) Wind Direction Sensor. This sensor should be aligned to true north and withstand a wind speed of 85 kts without damage.
(a) Range. 0º to 359º in azimuth.
(b) Threshold. 1 kt.
(c) Accuracy. Within ± 3º (RMSE).
(d) Resolution. To nearest 1º.
b. Performance Testing. If the manufacturer of the sensor has performed testing and it has been shown to meet the standards stated above, it is sufficient to submit documentation from the manufacturer establishing these facts, and further testing is not necessary. If testing is necessary, testing should be conducted in a calibrated wind tunnel. The ultrasonic sensor should be compared against a calibrated pitot-static tube or transfer reference standard traceable to the NIST.
3-7. TEMPERATURE SENSOR.
a. Performance Standards. The sensor should be thermally isolated to accurately measure environments below:
(1) Range. From -50ºC to +3ºC (-58ºF to +37.4ºF).
(2) Accuracy. 0.5ºC (1ºF) RMSE for entire range of the sensor, with maximum error of 1ºC (2 ºF).
(3) Resolution. Not greater than 0.5ºC (1ºF).
(4) Time Constant. Not greater than 2 minutes.
b. Performance Testing. If the manufacturer of the sensor has performed testing and it has been shown to meet the standards stated above, it is sufficient to submit documentation from the manufacturer establishing these facts, and further testing is not necessary. If testing is necessary, testing should be conducted in a temperature chamber.
3-8. PROVISIONS FOR LWES APPROVAL FOR SNOW.
a. Performance Standard for Identification of Snow. The following provision should be met if the system is to be approved for identification of snow: If precipitation of a LWE rate of at least 2.0 g/dm2/h (as determined by SAE Aerospace Recommended Practices (ARP) 5485 and 5945, a glycol precipitation rate pan measurement for a LWE sampling time of at least 5
Par 3-8 Page 10 minutes) has been detected and identified as snow by a human observer with credentials the FAA finds acceptable, the system should identify it as snow at least 95 percent of the time.
b. Performance Tests for Identification of Snow. A detailed protocol for performance of the tests should be included in the test plan described in Chapter 2. A minimum of 100 data points for an LWE sampling time of at least 5 minutes in snow varying from very light to heavy, and documented as described in the test protocol, should be collected to demonstrate compliance with performance standard in subparagraph 3-8a. At least 10 data points of very light (0 to 4 g/dm2/h), 15 data points of light (above 4 to 10 g/dm2/h), 30 data points of moderate (above 10 to 25 g/dm2/h) , and 15 data points of heavy (above 25 to 50 g/dm2/h) should be included in the data set. The data set should include a range of temperatures and wind speeds representative of those in which the system may be used.
NOTE: Heavy snowfall HOTs may be reported for snowfall rates up to 50 g/dm2/h. It is recommended that at least 15 data points during heavy conditions be submitted.
c. Performance Standards for Determination of LWE Rate for Snow. These provisions should be met if the system is to be approved for determination of LWE rate for snow:
(1) LWE rate measurements should be analyzed in reference to simultaneously recorded glycol precipitation rate pan measurements or in reference to simultaneously recorded measurements by another method the FAA finds acceptable. The testing period should be the same period used by the system operationally to determine HOT or CT, and cannot be less than 5 minutes.
(2) The LWES precipitation rate measurements should conform to the glycol precipitation rate pan measurements (refer to SAE ARPs 5485 and 5945) within the tolerances or alternate tolerance in Table 3-1 at a 95 percent confidence level:
TABLE 3-1. LWES SNOW RATE MEASUREMENT TOLERANCES LWES Rate g/dm2/h Acceptance Tolerance g/dm2/h Alternate Acceptable Tolerance From 2 to 10* -3.0 -4.0 g/dm2/h Above 10 to 25 -6.0 -40% Above 25 to 50 -14.0 * 0 g/dm2/h may be used as the lower limit of this range at the discretion of the applicant.
(3) If the system also meets the stated criteria in Table 3-1 when a minus (-) sign is replaced with a plus or minus (±) sign, then the system meets not only the stated (1-sided) criteria, but also a more stringent (2-sided) criteria. In its response to the report and data submission by the manufacturer, the FAA can provide a statement that this more stringent standard has been met.
Par 3-8 Page 11
d. Performance Tests for Determination of LWE Rate for Snow. A detailed protocol for performance of the tests should be included in the test plan required in paragraph 2-5.
A minimum of 100 data points for the LWE sampling time in snow varying from very light to heavy, and documented as described in the test protocol, should be collected to demonstrate compliance with the performance standard in subparagraph 3-8c. At least 10 data points of very light (0 to 4 g/dm2/h), 15 data points of light (above 4 to 10 g/dm2/h), 30 data points of moderate (above 10 to 25 g/dm2/h) , and 15 data points of heavy (above 25 to 50 g/dm2/h) should be included in the data set. The data set should include a range of temperatures and wind speeds representative of those in which the system may be used.
NOTE: Heavy snowfall HOTs may be reported for snowfall rates up to 50 g/dm2/h. It is recommended that at least 15 data points during heavy conditions be submitted.
e. Determination of HOT or CT Using LWE Rate and Temperature. These provisions should be met if the system is to be approved for determination of HOT or CT using LWE rates for snow:
(1) The LWE rate used by the system for the purpose of computing fluid HOT or CT should be the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for each range. Notwithstanding, the precipitation rate used by the LWES for the purpose of computing fluid HOT or CT must not be less than
2.0 g/dm2/h.
(2) The temperature used by the system for the purpose of computing fluid HOT or CT must be no higher than 0°C (32° F). Any temperatures reported by the system in excess of 0°C should be set to 0°C for the purpose of computing fluid HOT or CT.
(3) HOT or CT determinations must not be provided in snow conditions exceeding 50 g/dm2/h. This threshold is to be compared to the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for the appropriate range. In accordance with the currently accepted practice for HOT table development, HOT and CT determinations from the LWES for Type II and IV anti-icing fluids must be capped at 3 hours for snow.
(4) In accordance with the currently accepted practice for HOT table development, HOT and CT determinations from the LWES for Type II and IV anti-icing fluids must be capped at 3 hours for snow.
NOTE: If the accepted practice for HOT table development with respect to HOT caps is modified, these modifications should be applied to HOTDS and CTDS criteria.
3-9. PROVISIONS FOR LWES APPROVAL FOR FREEZING RAIN.
a. Performance Standards for Identification of Freezing Rain. The following provision should be met if the system is to be approved for identification of freezing rain. If precipitation of an LWE rate of at least 2.0 g/dm2/h (as determined by SAE ARPs 5485 and 5945, a
Par 3-9 Page 12 precipitation rate pan measurement for an LWE sampling time of at least 5 minutes) has been detected and identified as freezing rain by a human observer with appropriate credentials, the system should identify it as freezing rain at least 95 percent of the time.
b. Performance Tests for Identification of Freezing Rain. A detailed protocol for performance of the tests should be included in the test plan described in Chapter 2. A minimum of 25 data points for an LWE sampling time of at least 5 minutes in freezing rain, where at least 20 data points are in the light range below 25 g/dm2/h, and documented as described in the test protocol, should be collected to demonstrate compliance with the performance standard in subparagraph 3-9a. The data set should include a range of temperatures and wind speeds representative of those in which the system may be used.
c. Performance Standards for Determination of LWE Rate for Freezing Rain. These provisions should be met if the system is to be approved for determination of LWE rate for freezing rain:
(1) LWE rate measurements should be analyzed in reference to simultaneously recorded precipitation rate pan measurements or in reference to simultaneously recorded measurements by another method. The testing period should be the same period used by the system operationally to determine HOT or CT, and cannot be less than 5 minutes.
(2) The LWES precipitation rate measurements should conform to the precipitation rate pan measurements (ref.: SAE ARPs 5485 and 5945) or other measurements acceptable to the FAA, within the tolerances or alternate tolerances in Table 3-2 at a 95 percent confidence level:
TABLE 3-2. THE LWES FREEZING RAIN RATE MEASUREMENT TOLERANCES LWES Rate g/dm2/h Acceptable Tolerance g/dm2/h Alternate Acceptable Tolerance From 2 to 13* -3.0 -4.0 g/dm2/h Above 13 to 25 -6.0 -40% Above 25 -14.0 * 0 g/dm2/h may be used as the lower limit of this range at the discretion of the applicant.
(3) If the system also meets the stated criteria in Table 3-2 when a minus (-) sign is replaced with a plus or minus (±) sign, then the system meets not only the stated (1-sided) criteria, but also a more stringent (2-sided) criteria. In its response to the report and data submission by the manufacturer, the FAA can provide a statement that this more stringent standard has been met.
d. Performance Tests for Determination of LWE Rate for Freezing Rain. A detailed protocol for performance of the tests should be included in the test plan required in paragraph 2-5. A minimum of 25 data points for the LWE sampling time in freezing rain, mainly in the light range, and documented as described in the test protocol, should be collected to demonstrate compliance with the performance standard in subparagraph 3-9c.
Par 3-9 Page 13
e. Determination of HOT or CT Using LWE Rate. These provisions should be met if the system is to be approved for determination of HOT or CT using LWE rates for freezing rain:
(1) The LWE rate used by the system for the purpose of computing fluid HOT or CT should be the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for each range. Notwithstanding, the precipitation rate used by the LWES for the purpose of computing fluid HOT or CT must not be less than
2.0 g/dm2/h.
(2) HOT or CT determinations must not be provided in freezing rain conditions exceeding 25 g/dm2/h. This threshold is to be compared to the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for the appropriate range.
(3) In accordance with the currently accepted practice for HOT table development, HOT and CT determinations from the LWES for Type II and IV anti-icing fluids must be capped at 3 hours for freezing rain.
NOTE: If the accepted practice for HOT table development with respect to HOT caps is modified, these modifications should be applied to HOTDS and CTDS criteria.
3-10. PROVISIONS FOR LWES APPROVAL FOR FREEZING DRIZZLE.
a. Performance Standard for Identification of Freezing Drizzle. The following provision should be met if the system is to be approved for identification of freezing drizzle.
If precipitation of an LWE rate of at least 2.0 g/dm2/h (as determined by SAE ARPs 5485 and 5945, a precipitation rate pan measurement for an LWE sampling time of at least 5 minutes) has been detected and identified as freezing drizzle by a human observer with appropriate credentials, the system should identify it as freezing drizzle at least 80 percent of the time.
b. Performance Tests for Identification of Freezing Drizzle. A detailed protocol for performance of the tests should be included in the test plan described in Chapter 2. A minimum of 25 data points for an LWE sampling time of at least 5 minutes in freezing drizzle, distributed over the light, moderate, and heavy range, and documented as described in the test protocol, should be collected to demonstrate compliance with performance standard in subparagraph 3-10a. The data set should include a range of temperatures and wind speeds representative of those in which the system may be used.
c. Performance Standards for Determination of LWE Rate for Freezing Drizzle.
These provisions should be met if the system is to be approved for determination of LWE rate for freezing drizzle:
(1) LWE rate measurements should be analyzed in reference to simultaneously recorded precipitation rate pan measurements or in reference to simultaneously recorded measurements by another method. The testing period should be the same period used by the system operationally to determine HOT or CT, and cannot be less than 5 minutes.
Par 3-10 Page 14
(2) The LWES precipitation rate measurements should conform to the precipitation rate pan measurements (refer to SAE ARPs 5485 and 5945) or other measurements acceptable to the FAA, within the tolerances or alternate tolerances in Table 3-3 at a 95 percent confidence level:
TABLE 3-3. THE LWES FREEZING DRIZZLE RATE MEASUREMENT
TOLERANCES LWES Rate g/dm2/h Acceptable Tolerance g/dm2/h Alternate Acceptable Tolerance From 2 to 5* -3.0 -4.0 g/dm2/h Above 5 to 13 -6.0 -40% * (0 g/dm2/h may be used as the lower limit of this range at the discretion of the applicant.)
(3) If the system also meets the stated criteria in Table 3-3 when a minus (-) sign is replaced with a plus or minus (±) sign, then the system meets not only the stated (1-sided) criteria, but also a more stringent (2-sided) criteria. In its response to the report and data submission by the manufacturer, the FAA can provide a statement that this more stringent standard has been met.
d. Performance Tests for Determination of LWE Rate for Freezing Drizzle. A detailed protocol for performance of the tests should be included in the test plan required in paragraph 2-5. A minimum of 25 data points for the LWE sampling time in freezing drizzle documented as described in the test protocol should be collected to demonstrate compliance with performance standard in subparagraph 3-10c.
e. Determination of HOT or CT Using LWE Rate. These provisions should be met if the system is to be approved for determination of HOT or CT using LWE rates for freezing drizzle:
(1) The LWE rate used by the system for the purpose of computing fluid HOT or CT should be the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for each range. Notwithstanding, the precipitation rate used by the LWES for the purpose of computing fluid HOT or CT must not be less than
2.0 g/dm2/h.
(2) HOT or CT determinations must not be provided in freezing drizzle conditions exceeding 13 g/dm2/h. This threshold is to be compared to the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for the appropriate range.
(3) In accordance with the currently accepted practice for HOT table development, HOT and CT determinations from the LWES for Type II and IV anti-icing fluids must be capped at 3 hours for freezing drizzle.
Par 3-10 Page 15 NOTE: If the accepted practice for HOT table development with respect to HOT caps is modified, these modifications should be applied to HOTDS and CTDS criteria.
3-11. PROVISIONS FOR LWES APPROVAL FOR SUPERCOOLED LARGE DROPS (SLD).
NOTE: These provisions pertain to a system that can identify Supercooled Large Drops (SLD) (i.e., freezing rain or freezing drizzle), but cannot distinguish between freezing rain and freezing drizzle.
a. Performance Standards for Identification of SLD. The following provision should be met if the system is to be approved for identification of SLD. If precipitation of an LWE rate of at least 2.0 g/dm2/h (as determined by a precipitation rate pan measurement for an LWE sampling time of at least 5 minutes) has been detected and identified as SLD by a human observer with credentials the FAA finds, the system should identify it as SLD at least 95 percent of the time.
b. Performance Tests for Identification of SLD. A detailed protocol for performance of the tests should be included in the test plan described in Chapter 2. A minimum of 25 data points for an LWE sampling time of at least 5 minutes in SLD, distributed over freezing drizzle and light freezing rain, and documented as described in the test protocol, should be collected to demonstrate compliance with performance standard in subparagraph 3-11a. The data set should include a range of temperatures and wind speeds representative of those in which the system may be used.
c. Performance Standards for Determination of LWE Rate for SLD. These provisions should be met if the system is to be approved for determination of LWE rate for SLD:
(1) LWE rate measurements should be analyzed in reference to simultaneously recorded precipitation rate pan measurements or in reference to simultaneously recorded measurements by another method acceptable to the FAA. The testing period should be the same period used by the system operationally to determine HOT or CT, and cannot be less than 5 minutes.
(2) The LWES precipitation rate measurements should conform to the precipitation rate pan measurements (refer to SAE ARPs 5485 and 5945) or other measurements the FAA finds acceptable, within the tolerances or alternate tolerances in Table 3-4 at a 95 percent confidence level:
Par 3-11 Page 16 TABLE 3-4. THE LWES SLD RATE MEASUREMENT TOLERANCES LWES Rate g/dm2/h Acceptance Tolerance g/dm2/h Alternate Acceptance Tolerance From 2 to 13* -3.0 -4.0 g/dm2/h Above 13 to 25 -6.0 -40% Above 25 -14.0 * 0 g/dm2/h may be used as the lower limit of this range at the discretion of the applicant.
(3) If the system also meets the stated criteria in Table 3-4 when a minus (-) sign is replaced with a plus or minus (±) sign, then the system meets not only the stated (1-sided) criteria, but also a more stringent (2-sided) criteria. In its response to the report and data submission by the manufacturer, the FAA can provide a statement that this more stringent standard has been met.
d. Performance Tests for Determination of LWE Rate for SLD. A detailed protocol for performance of the tests should be included in the test plan required in paragraph 2-5.
A minimum of 25 data points for the LWE sampling time in SLD, distributed over freezing drizzle and light freezing rain, and documented as described in the test protocol, should be collected to demonstrate compliance with the performance standard in paragraph 3-11c.
e. Determination of HOT or CT Using LWE Rate. These provisions should be met if the system is to be approved for determination of HOT or CT using LWE rates for SLD:
(1) HOT or CT determinations for SLD should be based on computing the associated HOT or CT for freezing rain as specified in subparagraph 3-9e(1) and freezing drizzle as specified in subparagraph 3-10e(1) and using the resulting lower value.
(2) HOT or CT determinations must not be provided in SLD conditions exceeding 25 g/dm²/h. This threshold is to be compared to the LWE rate determined by the system plus the tolerance (within a 95 percent confidence level) that has been demonstrated for the appropriate range.
(3) In accordance with the currently accepted practice for HOT table development, HOT and CT determinations from the LWES for Type II and IV anti-icing fluids must be capped at 3 hours for SLD.
NOTE: If the accepted practice for HOT table development with respect to HOT caps is modified, these modifications should be applied to HOTDS and CTDS criteria.
Par 3-12 Page 17 (and 18)
3-12. LWES LATENCY. The LWES should generate an updated HOT or CT every 10 minutes or less.
a. LWES Inputs. The inputs the system uses to generate a HOT/CT are Air Temperature, Precipitation Type, LWE, and Wind Speed (optional) and should not be more than 10 minutes old. These inputs should also be reported every 10 minutes along with the updated HOT/CT.
b. Onset of Winter Precipitation. The LWES should determine and report a HOT/CT within 10 minutes of the onset of a winter precipitation event as identified in paragraphs 3-8 through 3-11c.