[THIS PAGE INTENTIONALLY LEFT BLANK]
XII. ICAO MANUAL APPENDIX 2 APPENDIX 2. SMOKE EMISSION EVALUATION Supplemental information The smoke measurement standard described in IC AO Annex 16, Volume II, Second Edition, 1993, Appendix 2, is an indirect measure of smoke plume visibility which is obtained by using a filter to trap smoke particles contained in a predetermined ma ss of exhaust gas, and measuring the loss of reflectance, i.e., degree of staining, of this filter rela tive to the absolute reflectance of the filter when clean or free of stain. It is a reasonably repeatabl e, simple procedure, but does not provide a direct measure of either plume visibility or any properties of the smoke particles, size, size distribution or mass. The SAE estimates the practically achievab le precision of the smoke number by this method to be within ±3 SN (smoke numbers). The equival ent SAE procedure is found in Aerospace Recommended Practice ARP1179, "Aircraft Gas Turbine Engine Exhaust Smoke Measurement".
1. INTRODUCTION AND DEFINITIONS Note.- The procedures specified here are concerned with the acquisi tion of representative exhaust samples and their transmission to, and analysis by, the emissions measuring system.
1.1 Variations in the procedure contained in this Appendix shall only be allowed after prior application to and approval by the certificating authority.
1.2 Where the following expressions and symbols are used in this Appendix, they have the meanings ascribed to them below:
Sampling reference size. The sample mass, 16.2 kg/m 2 of stained filter area, which if passed through the filter material results in a change of reflectance which gives a value of the SN parameter.
Sampling size. A chosen exhaust sample, the magnitude of whose mass (expressed in kilograms per square metre of stained filter surface area) lies in the range prescribed in
2.5.3 h) of this Appendix which, when passed through the filter material, causes a change in reflectance yielding a value for the SN' parameter.
Sampling volume. The chosen sample volume (expressed in cubic metres) whose equivalent mass, calculated as indicated in 3 of this Appendix, conforms to the above definition of sampling size.
SN. Smoke Number; Dimensionless term quantifying smoke emission level based upon the staining of a filter by the reference mass of exhaust gas sample, and rated on a scale of 0 to 100 (see 3 of this Appendix).
SN'. Smoke Number obtained from an individual smoke sample, not necessarily of the reference size, as defined in 3 of this Appendix.
W.. Mass of individual exhaust gas smoke sample, in kilograms, calculated from the measurements of sample volume, pressure and temperature (see 3 of this Appendix).
2. MEASUREMENT OF SMOKE EMISSIONS
2.1 Sampling probe for smoke emissions
a) The probe shall be made of stainless steel. if a mixing probe is used, all sampling orifices shall be of equal diameter.
b) The probe design shall be such that at least 80 per cent of the pressure drop through the probe assembly is taken at the orifices.
c) The number of sampling orifices shall not be less than 12.
d) The sampling plane shall be as close to the engine exhaust nozzle exit plane as permitted by considerations of engine performance but in any case shall be within 0.5 nozzle diameters of the exit plane.
e) The applicant shall provide evidence to t he certificating authority, by means of detailed traverses, that the proposed probe design and position does provide a representative sample for each prescribed power setting.
Supplemental information The following diagram, Figure A2-1, is an artists rendition of an averaging probe in a cruciform configuration, with the minimum number of 12 orificies located on centers of equal area. While typical of averaging probes the actual design is not limited to a cruciform configuration. As with any design, this configuration would have to be proven by means of a detailed traverse.
Supplemental Figure A2-1: 12 Point Cruciform Core Averaging Rake Page 123
Sampling emissions from aircraft gas turbine engines is difficult and has been under continual study and development. Lacking a definition of representa tive or constraints on a traversing pattern or protocol, some technical guidance can be obtained from looking at the methods described in early versions of SAE ARP 1179 and the Federal Regulations , developed primarily from experience within the industry in testing turbojet engines.
ARP 1179 stated that the tip of the sampling pr obe must be located between the engine exit plane and 6 inches downstream of the plane. The probe was to be oriented normal to this plane.
An initial traverse was required to arrive at an appropriate probe location for subsequent testing and a minimum traverse pattern was defined. This was a requirement for any given engine model or installation and at each engine power setting. The traverse was to be made radially at two locations 90 degrees apart, with 5 equal area sampling locations on each of the two radii. For each power setting, the point nearest the arithmetic average SN of the 10 points, was to be identified and entered as a permanent record of the particular engine or installation. All subsequent samples for that particular power setting were to be taken at this location. The sampling traverse needs to be repeated after any engine modification involving possible aerodynamic changes.
Supplemental Figure A2-2: ARP1179 Early Traverse Pattern In the first version of 40 CFR Part 87, Federal Register, Vol. 38, No. 136, July 17, 1973, the sampling protocol required a minimum of 12 sampling poi nts. Either mixing or individual probes were acceptable. At least three different radial positions were to be used in each of four sampling quadrants. If the minimum of 12 sampling points were used, the points in circumferentially adjacent sampling areas were to be separated in any direction by a distance not less than 0.1 tailpipe radius or
0.1 annular height, as applicable; if the number of sampling points (n) was greater than 12, they were to be equal in number in each quadrant or sector, and the minimum separations were to be reduced by a factor of 12/n.
The axial sampling plane had to be as close to the plane of the exit nozzle as engine performance parameters would permit, but in any case had to be within one exit nozzle diameter of the exit plane.
In all cases, the probe had to be designed to obt ain a representative sample over the area of the entire exhaust nozzle, on mixed fan engines, no nmixed fan engines, turbojets, turboprops, and auxiliary power engines. Finally, the multipoint probe had to be designed to minimize the errors due to pollutant stratification, whether the stratification wa s due to combustor design, mixing, lack of mixing, or engine design such as mixing of fan and core air.
In Federal Register Vol. 43, No. 58, March 24, 1978, this sampling protocol was modified to include a specific angular reference, i.e., if the minimu m of 12 sampling points were used, the points in circumferentially adjacent sampling areas were to be separated by at least 80 ° angular displacement.
The most recent version of ARP 1179 has the axial location of the sampling plane as close to the exit nozzle as engine performance parameters permit, but in any case within 0.5 exit nozzle diameter of the exit plane.
The sampling points are to be arranged over the ex haust nozzle exit area for straight turbojet, turboprop, turboshaft, and mixed flow (or confluent flow) fan engines, and over the core nozzle exit area for nonmixed fan engines. In order to prom ote uniformity of smoke measurements, a specific probe design must be standardized for use with a given type or series of engines, and the suitability of this probe must be demonstrated, by means of detailed traverse measurements in the sampling plane.
Sampling for smoke has been and continues to be a difficult problem to address and manage. If at all possible, samples should be taken from the core and analytically extended to the total exhaust plume.
As with gaseous emissions, consensus seems to be that averaging probes sampling points should be located on centers of equal area rather than equal spacing. The number of sampling orifices should probably be more than the 12 minimum required.
2.2 Sampling line for smoke emissions
2.2.1 The sample shall be transferred from the probe to the sample collection system via a line of 4.0 to 8.5 mm inside diameter taking the shortest route practicable which shall in no case be greater than 25 m. The line temper ature shall be maintained at a temperature between 600C and 1750C with a stability of ± l50C, except for the distance required to cool the gas from the engine exhaust temperature down to the line control temperature.
2.2.2 Sampling lines shall be as "straight thro ugh" as possible. Any necessary bends shall have radii which are greater than 10 times the inside diameter of the lines. The material of the lines shall be such as to discourage build-up of particulate matter or static electricity.
Note. — Stainless steel, copper or carbon-lo aded grounded polytetrafuoroethylene (PTFE) meet these requirements.
Supplemental information The line from the probe to the sampling system sh ould be heated to prevent water from condensing.
It should be short and relatively straight to discour age loss of particulate matter to the walls and at bends. It should be of reasonably small diameter to keep the flow rate high, minimizing particle dropout and wall loss. It should be constructed of a material that discourages static buildup which can lead to particle loss. If carbon-loaded polytetrafluoroethylene (PTFE - Teflon) is used as a line material it is important that the sample gas have a chance to cool before entering the sample line to protect the Teflon from burning.
2.3 Smoke analysis system Note. — The method prescribed herein is based upon the measurement of the reduction in reflectance of a filter when stained by a given mass flow of exhaust sample.
The arrangement of the various components of th e system for acquiring the necessary stained filter samples shall be as shown schematically in Figure 2-1. An optional bypass around the volume meter may be installed to facilitate meter reading. The major elements of the system shall meet the following requirements:
a) sample size measurement: a wet or dry positive displacement volume meter shall be used to measure sample volume to an accuracy of ± 2 per cent. The pressure and temperature at entry to this meter shall also be measured to accuracies of 0.2 per cent and ± 2 ºC respectively;
b) sample flow rate measurement: the sample flow rate shall be maintained at a value of 14 ± 0.5 L/min and the flowmeter for this purpose shall be able to make this measurement Page 125
with an accuracy of ± 5 per cent;
c) filter and holder: the filter holder shall be constructed in corrosion-resistant material and shall have the flow channel configuration shown in Figure 2-1. The filter material shall be Whatman type No. 4, or any equivalent approved by the certificating authority;
d) valves: four valve elements shall be provided as indicated in Figure 2-1:
1) valve A shall be a quick-acting, full-flow, flow diverter enabling the incoming sample to be directed through the measuring filter or around the bypass circuits of shut off;
2) Note. — Valve A may, if necessary, consis t of two valves interlocked to give the requisite function.
2) valves B and C shall be throttling valves used to establish the system flow rate;
3) valve D shall be a shut-off valve to enable the filter holder to be isolated;
all valves shall be made of corrosion-resistant material;
e) vacuum pump: this pump shall have a no-flow vacuum capability of -75 kPa with respect to atmospheric pressure; its full-flow rate shall not be less than 28 L/min at normal temperature and pressure;
f) temperature control: the incoming sample line through to the filter holder shall be maintained at a temperature between 60 °C and 175 °C with a stability of ± 15 °C;
Note. — The objective is to prevent water condensation prior to reaching the filter holder and within it.
g) If it is desired to draw a higher sample flow rate through the probe than through the filter holder, an optional flow splitter may be located between the probe and valve A (Figure 2
1), to dump excess flow. The dump line shall be as close as possible to probe off-take and shall not affect the ability of the sampling system to maintain the required 80 per cent pressure drop across the probe assembly. The dump flow may also be sent to the CO 2 analyser of complete emissions analysis system.
h) If a flow splitter is used, a test shall be conducted to demonstrate that the flow splitter does not change the smoke level passing to the filter holder. This may be accomplished by reversing the outlet lines from the flow splitter and showing that, within the accuracy of the method, the smoke level does not change.
i) leak performance: the sub-system sha ll meet the requirements of the following test:
1) clamp clean filter material into holder,
2) shut off valve A, fully open valves B, C and D.
3) run vacuum pump for one minute to reach equilibrium conditions;
4) continue to pump and measure the volume flow through the meter over a period of five minutes. This volume shall not exceed 5 L (referred to normal temperature and pressure)and the system s hall not be used until this standard has been achieved.
j) reflectometer: the measurements of the refl ectance of the filter material shall be by an instrument conforming to the American Nation al Standards Institute (ANSI) Standard No.
PH2.17/1977 for diffuse reflection density. T he diameter of the reflectometer light beam on the filter paper shall not exceed D/2 nor be less than D/10 where D is the diameter of filter stained spot as defined in Figure 2-1.
Supplemental information Maintaining the diameter of the reflectometer light beam on the filter paper between these limits insures averaging over a reasonable area of the filter. This is most important for stained spots which are not uniform in consistency of stain due to edge effects at the boundary of the stain and graininess within the spot particularly for lightly loaded filters, i.e., low SN.
Figure A2-3. Smoke analysis system
2.4 Fuel specifications The fuel shall meet the specifications of Appendi x 4. Additives used for the purpose of smoke suppression (such as organo-metallic compounds) shall not be present.
Supplemental information This fuel specification requirement is preempted by FAR 34 Subpart H, §34.81, Fuel specifications, which state “fuel having specifications as prov ided in §34.61 shall be used in smoke emission testing.“ Page 127
2.5 Smoke measurement procedures
2.51 Engine operation
2.5.1.1 The engine shall be operated on a stat ic test facility which is suitable and properly equipped for high accuracy performance testing.
2.5.1.2 The tests shall be made at the power settings approved by the certificating authority. The engine shall be stabilized at each setting.
2.5.2 Leakage and cleanliness checks No measurements shall be made until all sample transfer lines and valves are warmed up and stable. Prior to a series of tests the system shall be checked for leakage and cleanliness as follows:
a) leakage check: isolate probe and close o ff end of sample line, perform leakage test as specified in 2.3g) with the exceptions that valve A is opened and set to "bypass", valve D is closed and that the leakage limit is 2 L. Restore probe and line interconnections;
b) cleanliness check:
1) open valves B, C and D;
2) run vacuum pump and alternately set valv e A to "bypass" and "sample" to purge the entire system with clean air for five minutes;
3) set valve A to "bypass";
4) close valve D and clamp clean filter material into holder. Open valve D;
5) set valve A to "sample" and reset back to "bypass" after 50 kg of air per square metre of filter has passed through the filter material;
6) measure resultant filter spot SN' as described in paragraph 3 of this Appendix.;
7) I f this SN' exceeds 3, the system sha ll be cleaned (or otherwise rectified) until a value lower than 3 is obtained.
The system shall not be used until the require ments of these leakage and cleanliness checks have been met.
Supplemental information Leakage checks are to ensure clean air does not leak into the system, thereby diluting the sample and lowering the SN. Cleanliness checks ensure t hat the sample will not be contaminated, resulting in higher than warranted SN values. To minimize contamination problems, if the probe cannot be moved out of the sampling stream during engine st art up, the probe and lines can be back pressured with a suitable clean gas.
2.5.3 Smoke measurement Smoke measurement shall be made independently of other measurements unless the smoke values so measured are significantly below the limiting values, or unless it can be demonstrated that the smoke values from simultaneous smoke and gaseous emission measurements are valid, in which case smoke measurements made be made simultaneously with gaseous emissions measurements. In all cases the bend radius re quirements for sampling lines detailed in 2.2.2 shall be strictly observed. The smoke analysis sub-system shall be set up and conform to the Page 128
specifications of 2.3. Referring to Figure 2-1, the following shall be the major operations in acquiring the stained filter specimens:
a) during engine operations with the probe in position, valve A shall not be placed in the noflow condition, otherwise particulate buildup in the lines might be encouraged;
b) set valve A to "bypass" close valve D and clamp clean filter into holder. Continue to draw exhaust sample in the bypass setting for at least five minutes while the engine is at or near to the requisite operating mode, valve C being set to give a flow rate of 14 ± 0.5 L/min;
c) open valve D and set valve A to "sample", use valve B to set flow rate again to value set in b);
d) set valve A to "bypass" and close valve D, clamp clean filter material into the holder;
Supplemental procedure As a result of the manufacturing process, What man type No. 4 filters will be found to be slightly different, one side to the other, when measured using a reflectometer. This difference can approach 2 SN in value. During the filter cutting process, (filter disks are cut from larger sheets), the filters take on a slight set so there are distinct concave and c onvex sides. Filters seem to be cut, packaged and shipped in batches which preserves the top/bottom- concave/convex relationship. To eliminate unnecessary errors in measurement, it is advisable to load the filters in the filter holder in a consistent manner, i.e., always concave or convex side up. This should be kept in mind as well when determining the absolute reflectance of the clean filter, e.g., if the filter is placed in the filter holder concave side up, that is facing the sampling stream, then the absolute reflectance of the concave side is the reference measurement to be made.
e) when the engine is stabilized on condition, a llow one minute of sample flow with settings as at d);
f) open valve D, set valve A to "sample", rese t flow rate if necessary, and allow chosen sample volume (see h) to pass, before setting valve A back to "bypass" and close valve D;
g) set aside stained filter for analys is, clamp clean filter into holder;
h) the chosen sample sizes shall be such as to be within the range of 12 kg to 21 kg of exhaust gas per square metre of filter, and sh all include samples which are either at the value of 16.2 kg of exhaust gas per square metre of filter or lie above and below that value. The number of samples at each engine operating condition shall not be less than 3 and e) to g) shall be repeated as necessary.
3. CALCULATION OF SMOKE NUMBER
FROM MEASURED DATA The stained filter specimens obtained as outlined in 2.5.3 shall be analysed using a reflectometer as specified in 2.3. The backing material used shall be black with an absolute reflectance of less than 3 per cent. The absolute reflectance reading R s of each stained filter shall be used to calculate the reduction in reflectance by SN' = 100(1-Rs/Rw) Where Rw is the absolute reflectance of clean filter material.
Supplemental information Page 129
Reflectometers sometimes come equipped with a pseudo 'black body' cup like device which is used to set the meter zero response. It is not a good idea to use a device like this as backing material, although it does have an absolute reflectance of le ss than 3%. A solid surface, which applies light and equal pressure over the filter area to be me asured, and which meets t he less than 3% absolute reflectance requirement, is preferred.
The masses of the various samples shall be calculated by W = 0.348 PV/T x 10-2 (kg) Where P and T are, respectively, the sample pressure in pascals and the temperature in kelvin, measured immediately upstream of the volume meter. V is the measured sample volume in cubic metres.
For each engine condition in the case that the sample sizes range above and below the reference value, the various values of SN' and W shall be plotted as SN' versus log W/A, where A is the filter stain area (m 2). Using a least squares straight line fit, the value of SN' for W/A=16.2 kg/m 2 shall be estimated and reported as the Smoke Number (SN) for that engine mode. Where sampling at the reference size value only is em ployed, the reported SN shall be the arithmetic average of the various individual values of SN'.
Supplemental information Experience has shown that plotting the set of meas ured SN’s against the logarithm of the equivalent W/A’s results in a straight line. ICAO requires a least squares straight line fit to the SN’ vs. log W/A data, sometimes referred to as the regression of SN’ on log W/A.
The calculation is straightforward, and for the limited data sets involved with jet engine testing, can be performed on a hand held calculator or by means of a desktop computer spreadsheet. The values to be calculated are the straight line intercept, a, and slope, b.. The equations to be solved for the least squares estimates of the intercept and slope, SN’≡ yi and log(W/A)≡xi, are:
ya b + x= x − x y y∑ n ( i )( i − ) i=1b = n xx∑( i − ) 2 i=1 x bar and y bar are the means of xi and yi respectively and i refers to the sample number. For the testing required a simple table with columns for sample number i, SN’,W/A, log (W/A) and [log(W/A)] 2 in collecting and organizing the data and intermediate calculations.
For example, in the case where the minimum number of 3 smoke samples is taken a table might look like:
Table A2-1 TABLE FOR CALCULATING SN Sample number i (SN'I ) Log(W i/A) log(W i/A)*(SN'I) [log(Wi/A)]2 1 2 3 Average Sum Using this table and the previously calculated values of SN prime and W/A, it is a simple matter to calculate the estimated slope and intercept, and from them the least squares straight line. The reporting value of SN can be estimated from a plot of this straight line or calculated from the equation.
3 0 . EXHIBIT 1
Engine and Propeller Directorate And Office of Environment and Energy Inter-Office Management Plan for Aircraft Engine Fuel Venting and Exhaust Emissions The Engine and Propeller Directorate (EPD) and the Office of Environment and Energy (AEE) agree that an inter-office management plan is needed to strengthen FAA's Participation in, and oversight of aircraft engine fuel venting and exhaust emission certification. The Participating entities believe that the actions and procedures identified in this document provide the foundation for a Fuel Venting and Exhaust Emissions Procedures management plan.
The objective of this management plan is to define the roles and responsibilities of both entities, optimize the use of FAA resources, and provide adequate technical and procedural guidance for demonstrating compliance to 14 Code of Federal Regulations (CFR) Part 34. These actions are needed to:
1. Strengthen the FAA's surveillance of aircraft engine fuel venting and exhaust emissions certification;
2. Improve the application of technical guidance and procedural practices, and
3. Promote improved inter-office communication and cooperation.
Roles and Responsibilities: EPD and AEE agree that to achieve a step improvement in business practices between their respective organizations, the roles and responsibilities of Participating organizations need to be clearly defined. The following describes those roles and responsibilities:
XIII. Organization Role Responsibility Office of Environment and Energy Regulatory and Standardization
• Liaison with ICAO and CAEP
• Coordinates with EPA (40 CFR Part 87)
• Codifies fuel venting and exhaust emissions regulations (14 CFR 34)
• Processes applications for Exemptions (14 CFR Part 11)
• Audits and evaluates delegated functions
• Approves new or equivalent test procedures (14 CFR 34)
• Provides Policy and Guidance (AC-34) Page 132