Figure 2-10. Triple splitter example geometry as (a) single plane; (b) multi -plane; and (c) line diameter change
15.13. Appendix 7, 7.2.6: Switching between CO, HC, and NOx measurements and nvPM measurement while in stable operating condition While the engine is at a stable operating condition, sequential switching between CO, HC, and NOx measurements and nvPM measurement could be allowed.
15.14. Appendix 7, 7.2.6: Obtaining smoke measurements using the nvPM sampling system Smoke measurements may be obtained using the nvPM sampling system. The collection part (Section 1) of the nvPM sampling and measurement system meets the specifications in A16V2, Appendix 2 and the Splitter 1 assembly allows the GL, or another heated sampling line, to be used to measure smoke number as long as the sampling line requirements in Appendix 2 are met. If smoke number measurements are taken, it is recommended to obtain them sequentially with the nvPM measurements.
Installation of the smoke sampling line at Splitter 1 for certification measurements may increase the total sampling line length for smoke measurements by up to 8 m due to the collection part. In this case, the requirement in Appendix 2, 2.2.1 for a maximum sampling line length of 25 m may be exceeded, up to a maximum combined length of 33 m.
A smoke number reduction of less than 1 SN has been determined for an additional 8 m sampling line length by FOA3 modeling and direct measurements for smoke numbers up to 15. The FAA may consider granting an increase of sampling line length for smoke measurements. It is good practice to add 1 SN to the measured data as a conservative approach for determining the smoke number in such a case.
15.15. Appendix 7, 7.5.1: Examples of system flow rates and system layouts for system flow control Figure 2-11 shows a location example of two flow controllers installed in the make- up flow line to provide Module 3 system flow control. In the figure, Module 3 flow rate of 25 slpm is the sum of the nvPMmi (4 slpm), VPR (4 slpm), primary pump flow controller (14 slpm) and CO2 pump flow controller (3 slpm).
Page 36 of 57
Section 3 flow rate 25 slpm 4 slpm 4 slpm 3 slpm Splitter 2 Filter Filter Flow controller Flow controller Pump 14 slpm VPR (DF )2 nvPM mass instrument nvPMmi Primary pump CO analyser 2 CO[ ] 2 dil1 Figure 2-11. Example of system flow rates that ensure Module 3 flow rate satisfies 25 ± 2 slpm Figure 2-12 shows examples of sampling layout options for performing nvPM sampling system fl ow control and [CO2]dil1 measurement for determination of the first stage dilution factor DF1.
Filter Filter Filter Flow controller Flow controller Flow controller Flow controller Pump Pump Pressure control valve Primary pump Primary pump Primary pump CO analyser 2 CO[ ] 2 dil1 CO analyser 2 CO[ ] 2 dil1 CO analyser 2 CO[ ] 2 dil1 Figure 2-12. Examples of sampling layout for nvPM sampling system flow control at first stage dilution measurement The sample gas measured downstream of Diluter 1 can be considered “dry” as it will consist of at least 88 per cent dry air or dry nitrogen. The diluted sample water content is negligible and consequently there is no “wet” correction required for the CO2 measurement downstream of Diluter 1.
15.16. Appendix 7, 8: Using thermal optical transmittance (TOT) analysis to determine elemental carbon mass as a surrogate for black carbon mass for calibration purposes Non-volatile PM (nvPM) is defined as those particles present at the aircraft engine exit plane which do not volatilize when heated to a temperature of 350°C. nvPM consists mainly of nanometre- size black carbon with trace amounts of ash and metallic particles emitted by aircraft engines under normal operating conditions. Total nvPM mass cannot be directly measured except by filter sampling and analysis which is expensive and time consuming. Therefore, the on-line measurement of black carbon mass concentration is considered the most appropriate method for representing the nvPM emissions from aircraft engines.
Page 37 of 57
For calibration purposes, however, the mass of elemental carbon (EC), determined by thermal optical transmittance analysis, is used as a surrogate for black carbon mass since it was the most applicable method at the time the nvPM regulation was established.
15.17. Appendix 7, 8: Clarification of terms in Appendix 7 Table 2-3 provides a clarification of the terms used in Appendix 7 Section 8.
Table 2-3. Definition of select terms used in Appendix 7 Term Use Demonstrate Conformity To show that a nvPMmi meets an individual Appendix 7 Performance Specification (for a performance specification) Certificate Received after a nvPMmi has demonstrated conformity to all Performance Specifications Accuracy Part of Performance Specifications: Determine nvPMmi agreement with EC Mass concentration determined from TOT method, by way of linear regression Verification Part of Performance Specifications: Verifying the combination of instrument and annual calibration source for nvPM mass measurement on an aircraft gas turbine engine exhaust Annual Calibration Adjust nvPMmi calibration factor to agree with EC mass concentration determined from the TOT method
15.18. Appendix 7, 8.2.1: Thermal optical transmission (TOT) method The thermal optical transmission (TOT) method can speciate EC from organic carbon (OC), and is performed by a thermal optical analyser with timed heating ramps and "cool-down" cycles. While the method measures all carbon species that evolve during the analysis and provides total carbon (TC) loading on the filter, the EC fraction is used for calibration. In the laboratory method, all carbon evolving from the filter is oxidized to form carbon dioxide (CO2), which is then reduced to methane (CH4), and measured using a flame ionization detector (FID), as described below.
A red light (wavelength of 670 nm) laser and a photocell are used to monitor transmittance of a filter punch of known area, which typically darkens as refractory OC chars during a non-oxidizing heat ramp and then lightens as the char burns off during an oxidizing heat ramp. Figure 2-13 illustrates a typical heat cycle of the TOT method. Note that Figure 2-13 is provided for illustration purposes and does not reflect the required temperature profile for the TOT method analysis cycle. Instrument software divides TC into OC and EC by evaluating the split time at which the transmittance of the filter returns to its original value from the beginning of the analysis. One should note that since the instrument uses light transmittance Page 38 of 57
through the filter to speciate the carbon types, any artefacts that remain on the filter (due to lack of complete charring etc.) can be determined as EC rather than OC during analysis. The range for this method is 1 to 105 µgC (micrograms carbon) per filter punch (usually about 1.5 cm 2). The limit of detection is about 0.2 µgC per cm2.
Figure 2-13 is a typical thermogram for a filter sample containing OC, carbonate carbon (CC) and EC. PC is carbon generated by pyrolysis. The curves indicated with the OC, CC, PC, EC and CH 4 labels are CH4 concentrations being measured by an FID for the laboratory TOT carbon analyser. The final peak is the methane calibration peak.
Figure 2-13. Typical thermogram for filter sample The optical component of the analyser is used to correct for pyrolysis of OC compounds to EC in order to avoid underestimation of OC and overestimation of EC. In order to ensure a proper OC/EC split, calibration of the internal oven temperature should be conducted according to the manufacturer’s specifications. The sample reflectance and transmittance are continuously monitored by a laser and a photo detector throughout the thermal cycle. When pyrolysis takes place, there is an increase in light absorption resulting in a decrease in reflectance and transmittance. Thus, by monitoring the reflectance/transmittance, the portion of the EC peak corresponding to pyrolyzed organic carbon can be Page 39 of 57
correctly assigned to the organic fraction. An example of the laser transmission amplitude (curve denoted with the “transmittance” label) is shown in Figure 2-13.
In the semi-continuous TOT analyser, a quartz filter is mounted directly in the instrument and samples are collected for the desired time. Once the collection is complete, the oven is purged with helium and a stepped-temperature ramp increases the oven temperature to 870°C, thermally desorbing organic compounds and pyrolysis products into a manganese dioxide (MnO2) oxidizing oven. As the carbon fragments flow through the MnO2 oven, they are quantitatively converted to CO2 gas. The CO2 is swept out of the oxidizing oven with the helium stream and measured directly by a self -contained non-dispersive infrared (NDIR) detector. A second temperature ramp (oven temperature to 930°C) is then initiated in a He/O2 gas stream and any EC is oxidized off the filter and into the oxidizing oven and NDIR. The EC is then detected in the same manner as the OC.
15.19. Appendix 7, 8.3: Procedure to demonstrate conformity to performance specifications The nvPM mass instrument (nvPMmi) is required to demonstrate conformity to the performance specifications in Table A7-5 to Annex 16, Volume II, Appendix 7 using a diffusion flame combustion aerosol source (DFCAS). Specifically, for the verification performance specification, as stated in Appendix 7, an aircraft turbine engine must be used as the combustion aerosol source.
A DFCAS is a device employing diffusion flame combustion using a given fuel that emits airborne particulate matter. A diffusion flame is a mode of combustion in which the fuel and oxidizer (air) are supplied separately to the combustion zone, where mixing by molecular diffusion takes place together with the combustion reactions. Hydrocarbon-fuelled diffusion flames produce soot (a form of non-volatile particulate matter) and organic aerosols (a form of volatile particulate matter). The DFCAS is a combination of the device to produce the flame and the fuel used.
Examples of a DFCAS Gaseous fuel — non-premixed methane inverted flame; non-premixed propane soot generator Liquid fuel — gas turbine combustor rig; gas turbine engine; diesel direct injection engine Diffusion flame combustion aerosol sources The following lists guidance on how to differentiate DFCASs.
Different models of laboratory aerosol generator are considered different DFCASs.
One aircraft engine type is considered different from another aircraft engine type.
A laboratory aerosol generator is considered a different DFCAS from an aircraft engine.
A change of operating settings on a DFCAS, which results in significant changes to the nvPMmi calibration factor (the relationship between instrument response and TOT reference) is considered a change of DFCAS.
Examples of DFCAS operating settings are changing the air-to-fuel ratio of a diffusion flame or sampling from different flame height locations within the DFCAS.
Using a DFCAS with a different serial number does not constitute a change of DFCAS.
It may be possible to show that the soot produced from one DFCAS is sufficiently similar to another that they can be considered the same DFCAS.
Page 40 of 57
15.20. Appendix 7, 8.3: Procedure to demonstrate conformity to nvPMmi performance specifications using a DFCAS The Technical Procedure below provides guidance for demonstrating conformity to nvPMmi performance specifications using a DFCAS.
The flowchart below describes the process followed by an engine manufacturer to perform a nvPM mass concentration measurement, how to verify a DFCAS, and how to demonstrate conformity to the performance specifications for a nvPMmi.
The flowchart describes 3 processes:
Testing Process (blue) This shows the procedure to measure nvPM mass concentration and perform a nvPMmi annual calibration Calibration with a new DFCAS (red) This shows the procedure to demonstrate that a new DFCAS can be used to calibrate the instrument. A new DFCAS may be a different particle source or the same particle source with a change in operational setting such that the calibration factor (the relationship between instrument response and TOT reference) may be significantly different.
Verifying a new DFCAS involves:
(i) Using the DFCAS to perform a nvPMmi calibration, applying the number of TOT filter tests specified in the “Calibration with a new diffusion flame combustion aerosol source” column of Appendix 7, Table A7-7.
(ii) Using an aircraft engine exhaust to demonstrate that the nvPMmi meets the verifi cation performance specification in Appendix 7, A7-5.
If a turbine source is used as the DFCAS for nvPMmi calibration, then a different aircraft engine type is used for the verification performance specification.
A new DFCAS may be used for nvPMmi calibration after it has been used for the verification performance specification for that nvPMmi make and model.
Instrument certificate (green) This shows the procedure to obtain a nvPMmi certificate in demonstrating conformity to the performance specifications in Appendix 7, Table A7-5. This process needs to be completed once for each nvPMmi make and model and again in the case of a change in hardware or software is made to the nvPMmi affecting the data acquisition and/or processing.
nvPMmi verification involves:
(i) Using a DFCAS to show that the nvPMmi meets the performance specifications (Repeatability, Zero drift, Linearity, Limit of Detection, Rise time and Sampling Interval) in Appendix 7, Table 7-5
(ii) Using a DFCAS calibration to show the nvPMmi meets the Accuracy performance specification in Appendix 7, Table 7-5, applying the number of TOT filter tests specified in the “Instrument Certificate” column of Appendix 7, Table A7-7.
(iii) Using an aircraft engine exhaust to demonstrate that the nvPMmi meets the verification performance specification of Appendix 7, A7-5 Page 41 of 57
If a turbine source is used as the DFCAS for nvPMmi calibration, then a different aircraft engine type is used for the verification performance specification.
Wish to perform nyPM test Are there changes Is this a new to the hardware and software ls the mvPMmi nuPMmi make of the nyMPmi affecting data ›No* calibration current?
or model?- acquiaition and processing?
Yes No Yes Perform calibration and demonstrate conformity to performence specifications (Table A7-5 Rows 1-7) Are there changes to the DFCAS device model, fuel type or operating settings?
(Table A7-5 Row 8) Perform Calibrate the nyPMmi Perform (Table A7-5 Row 8) verification beat (Table A7-7 3° column) with the new DFCAS calibration annual -Yes:
(Table A.7-8 Is verification Perform performance specification No nyPM test demonstrated?
Yes Receive DFCAS Receive nvPMmi documentation certificate [Documentation confiming verification confirming specifications] If the verification performance specification is not confirmed, the instrument is calibrated with a new DFCAS following the red (DFCAS verification) process above.
When the verification performance specification has been confirmed, the nvPMmi will receive a certificate, as stated in Appendix 7, paragraph 4.4.1.2.
The flow chart steps are detailed below:
Page 42 of 57
The engine manufacturer wishes to perform a nvPM test which requires using a nvPMmi.
a) If this is a new nvPMmi make and model, then the following steps are followed:
1) The nvPMmi demonstrates conformity to the performance specifications documented in Appendix 7, paragraph 8.1 and as stated in Appendix 7, Table 7-5, rows 2-8.
2) The nvPMmi is calibrated as stated in Appendix 7, paragraph 8.3.2.8, using the “instrument certificate (number of tests)” column (Appendix 7, Table A7-7).
3) The nvPMmi meets the verification performance specification requirements documented in Appendix 7, Table A7-5, ninth (last) row with the procedures stated in Appendix 7, paragraph 8.3.3.
4) If the nvPMmi does not meet the verification performance specification requirements, the procedure in steps 1), 2) and 3) above are repeated.
5) Once the nvPMmi meets the verification performance specification requirements, the nvPMmi will receive a certificate as stated in Appendix 7, paragraph 4.4.1.2.
6) Restart the process at step (a).
b) Else if there are hardware or software changes to the nvPMmi affecting data acquisition and/or processing then the following step is followed:
1) Move to Step (a.1)
c) Else if the nvPMmi calibration is not current, and the DFCAS model is new or if the DFCAS settings or fuel type are changed in a way that the calibration factor (the relationship between instrument response and TOT reference) may be significantly different, then the following steps are taken:
1) The nvPMmi is calibrated with the new DFCAS, as stated in Appendix 7, paragraph
8.3.2.8, using the “calibration with a new diffusion flame combustion aerosol source (number of tests)” column stated in Appendix 7, Table A7-7.
2) Move to step (a.3)
d) Else if the nvPMmi calibration is not current, and the DFCAS verification has been confirmed,
1) then an annual calibration of the nvPMmi is performed as stated in Appendix 7, paragraph 8.5.
2) Perform nvPM test
e) Else if the nvPMmi is an existing instrument make and model, approved and with a current calibration.
Perform nvPM test.
15.21. Appendix 7, 8.3.2: Measurement using a diffusion flame combustion aerosol source: conformity demonstration and annual calibration The calibration system for nvPM measurements should be located in a well-ventilated area and must contain at least a diffusion flame combustion source, an adjustable dilution system, a cyclone or other large particle remover, a plenum manifold system or a splitter, a quartz filter sampler or semi- continuous EC/OC analyser and the nvPM mass measurement instrument. Other diagnostic instruments may also be used. The sampling lines downstream of the plenum or splitter should be of matching material (stainless steel or carbon loaded polytetrafluoroethylene (PTFE)) and same geometry and flow rates to match Page 43 of 57
Table 2-4. B reakdown of sections and major components of calibration system Section I — PM combustion source II — Sample collection Sub-section I. Diffusion flame combustion aerosol source IIa. EC/OC analysis IIb. nvPMmi and diagnostic analyser(s) Major component(s) Burner: diffusion flame burner C1: 1 µm cut point stainless steel cyclone D1: diluter (air or N2) D2 :
Ejector venturi pump M1 : Mixing plenum M2 : Manifold sampling region Either:
FH1: stainless steel quartz filter ho lder for EC/OC determination using manual filter preparation and laboratory EC/OC analyser (shown) or Semi continuous EC/OC analyser (not shown) nvPMmi A-DIAG: optional diagnostic particle analyser particle losses in each sample line. An example calibration system is illustrated in Figure 2- 14 and the components identified in Table 2-4.
Figure 2-14. Schematic of an example instrument calibration system Page 44 of 57
In the system shown in Figure 2-14, soot particles are created in the diffusion flame, the characteristics of which will depend on the flow of fuel, air and diluent provided to the burner. Upon leaving the burner, the exhaust stream is diluted as necessary to meet the target soot concentrations. Large particles are then removed in the cyclone and the sample stream is distributed to the various instruments for analysis.
Care should be taken if using a splitter assembly. Best aerosol physics judgement should be used to avoid a potential for particle concentration bias between sampling legs. To reduce risk of a concentration bias between the filter sampler and instrument, plenum manifold systems are typically the preferred choice for instrument calibration systems, as long as the source aerosol is well mixed.
The exact instrumentation set-up should be based on the standard operating procedure for each instrument being calibrated. One example set-up using a splitter is shown in Figure 2-15. Another suitable system using an ejector, plenum and manifold is shown in Figure 2-16.
Figure 2-15. Schematic of an example instrument calibration system using a splitter Page 45 of 57
Figure 2-16. Schematic of an example instrument calibration system with a manifold located behind the cyclone
15.22. Appendix 7, 8.4.4: Determining the limit of detection (LOD) of the nvPMmi Equivalent methods of determining the limit of detection (LOD) are present in the technical literature and international standard. ISO 11843-1: 1997 Capability of Detection Part 1 Terms and Definitions provides the following:
• minimum detectable net concentration or amount true net concentration or amount of the analyte in the material to be analysed which will lead, with probability (1 - β), to the conclusion that the concentration or amount of the analyte in the analysed material is larger than that in the blank material International Union of Pure and Applied Chemistry (IUPAC) and ISO detection limits (minimum detectable amounts) are based on the theory of hypothesis testing and the probabilities of false positives α and false negatives β. The limit of detection is given as:
where Kα and α are associated with the one-sided tails of the distribution of the blank, with standard deviation σ0 corresponding to probability levels (1-α). Similarly, Kβ and β are associated with the onesided tails of the distribution of the limit of detection with standard deviation σD corresponding to probability levels (1- β). The latter is obtained at a low concentration near the detection limit. Often, it can be assumed that α=β and σ=constant, such that Kα= Kβ and σD=σ0. This assumption implies that the LOD varies with twice the critical level, LC, which is defined as:
Page 46 of 57
LC = Kασ0.
The values of α and β recommended by IUPAC and ISO documents are 0.05 each. If σ0 is estimated by s0, based on ν degrees of freedom, K can be replaced by Student’s t, such that:
LC = tν,0.95 s0 and LD = 2 tν,0.95 s0 which contributed to the formulation found in ISO 9169. Based on the assumptions that α=β and σ=constant, such that Kα= Kβ and σD=σ0, then by corollary sD=s0 and:
LD = 2 tν,0.95 sD.An equivalent procedure for determining the limit of detection is:
where:
YLOD,0.95 = Limit of detection at 95 per cent confidence interval tv,0.95 = The one-sided Student’s factor at 95 per cent confidence, degrees of freedom ν=n-1 where n is the number of measurements (repeats at a recommended 30 second averaging time) sLOD = The standard deviation of the measurements YLOD,j obtained by measuring at a concentration near the limit of detection over the averaging time.
15.23. Appendix 7, 8.5: Annual calibration using a DFCAS The change in nvPMmi annual calibration factor should not exceed the Accuracy performance specification (±10 per cent).
15.24. Appendix 7, 8.5.6: Context of “as found” performance for nvPM mass instrument calibration In the context of instrument calibration, “as found” is the condition in which the instrument was received by the laboratory performing the calibration prior to any alterations or adjustments. The “as found” condition is an indication of the operability of the instrument since its last calibration. Evaluating the instrument in this manner is useful for understanding instrument drift.
15.25. Appendix 7, 9.2.1: Context of “as found” performance for VPR calibration In the context of instrument calibration, “as found” is the condition in which the instrument was received by the laboratory performing the calibration prior to any alterations or adjustments. A certificate stating the “as found” values should be requested. For the VPR, dilution factors, particle penetration and particle removal efficiency should be reported. The “as found” condition is an indication of the state of operability of the instrument since its last calibration. Evaluating the instrument in this manner is useful for understanding instrument drift.
9.2.2 CPC calibration
15.26. Appendix 7, 9.2.2: Context of “as found” performance for CPC calibration In the context of instrument calibration, “as found” is the condition in which the instrument was received by the laboratory performing the calibration prior to any alterations or adjustments. A certificate stating the “as found” values should be requested. For a CPC, these values would include inlet flow, temperatures Page 47 of 57