STUDY ON
EMISSION CONTROL TECHNOLOGY FOR HEAVY-DUTY VEHICLES FINAL REPORT VOLUME 2 MEASUREMENT TECHNIQUES AND SAMPLING PROCEDURES FOR LOW LEVELS OF PARTICULATES
CONTRACT N° ETD/00/503430 Study prepared for the European Commission – DG ENTR (Enterprise)
Joint effort by MIRA Ltd, United Kingdom PBA, United Kingdom LAT/AUTh, Greece TU Graz, Austria TNO Automotive, Netherlands Vito, Belgium
May 2002 (Updated July 2002)
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
This part of the project was carried out by
Aristotle University Thessaloniki Department Mechanical Engineering Laboratory of Applied Thermodynamics POB 458 541 24 Thessaloniki Greece
Nikolas Kyriakis Tel. 30 310 996 083 Fax 30 310 996 019
[email protected]
Ilias Vouitsis Tel. 30 310 996 052 Fax 30 310 996 019
[email protected]
Zissis Samaras Tel. 30 310 996 014 Fax 30 310 996 019
[email protected]
Volume 2
Low Levels of Particulates
Page 2
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
0
CONTENTS
0
Contents .....................................................................................................................................3
1
Executive Summary ..................................................................................................................5
2
Introduction...............................................................................................................................7 2.1 Approach ...................................................................................................................................9
3
The existing and proposed pm mass measurement techniques ..........................................10 3.1 The Directive 1999/96/EC.......................................................................................................10 3.1.1 Dilution and Sampling System .................................................................................................... 10 3.1.2 Determination of Particulates ...................................................................................................... 17
3.2 The Draft ISO FDIS 16183 (Version 195E) ...........................................................................20 3.2.1 Dilution and Sampling System .................................................................................................... 20 3.2.2 Determination of Particulates ...................................................................................................... 21
3.3 US 2007 Federal Regulations..................................................................................................22 3.3.1 Dilution and Sampling System .................................................................................................... 22 3.3.2 Determination of Particulates ...................................................................................................... 26
4
Discussion of the provisions of the existing techniques .......................................................28 4.1 Fundamental considerations ....................................................................................................28 4.2 Sampling system configurations .............................................................................................33 4.2.1 Transient control capability of partial flow dilution systems....................................................... 35 4.2.2 Essential parameters that may influence PM measurement with partial flow systems................ 35 4.2.3 Correlation of partial flow systems with CVS ............................................................................. 38
4.3 Filter holder temperature .........................................................................................................42 4.4 Dilution air temperature and humidity ....................................................................................44 4.5 Particulate Sampling Probe .....................................................................................................45 4.6 Sample filter material ..............................................................................................................46 4.7 Number of sample filters.........................................................................................................46 4.8 Sample filter load ....................................................................................................................46 4.9 Stain diameter..........................................................................................................................47 Volume 2
Low Levels of Particulates
Page 3
EC-DG ENTR
5
Emission control technology for heavy-duty vehicles
ETD/00/503430
4.10
Sample filter conditioning................................................................................................47
4.11
Sample filter weighing .....................................................................................................48
4.12
Weighing conditions ........................................................................................................49
Alternative measurement methods for diesel particulate matter mass Measurements ...50 5.1 Horiba New Measurement Method .........................................................................................51 5.2 Real-time Measurement Techniques .......................................................................................53 5.2.1 Optical Methods........................................................................................................................... 53 5.2.2 Pressure Differential Technique .................................................................................................. 55 5.2.3 Fast-Response Flame Ionisation Detection.................................................................................. 55 5.2.4 Tapered Element Oscillating Microbalance (TEOM).................................................................. 58
6
Cost of pm measurement........................................................................................................69
7
Natural gas fuelled vehicles....................................................................................................71
8
Summary and Conclusions ....................................................................................................74
9
Acknowledgements .................................................................................................................76
10
References ................................................................................................................................77
11 Annex to Volume 2: Partial Flow Dilution Systems
Volume 2
Low Levels of Particulates
Page 4
EC-DG ENTR
1
Emission control technology for heavy-duty vehicles
ETD/00/503430
EXECUTIVE SUMMARY
There is potential to improve the current sampling and measurement procedures in the EU legislation for particulate matter (PM) emissions to allow repeatable measurements at the low emission levels of future engines.
The emerging new ISO/FDIS 16183 standard will specify procedures for partial flow dilution under transient testing conditions, which is currently not included in EU legislation. Although partial flow systems have been used satisfactorily for steady state tests, the known correlation issues have to be examined more closely for transient testing. Most of PM measurements conducted so far using partial flow systems have been found to agree within ±10% with the reference full dilution techniques, at around 20% of the capital cost. Better accuracy, at the level of the equivalency criterion (5% and below) defined by Directive 1999/96/EC, was recently reported by the EPA/ARB/EMA study.
Studies to identify optimal partial flow conditions show a small effect of the sample line length, while the current ISO specifications for sample line temperature, tunnel heating and sample line diameter seem to be acceptable. Variations in the dilution air temperature between 20 and 30oC is not considered to be significant. Statistically, the most significant parameter affecting the PM measured is the dilution ratio for both full flow and partial dilution systems. It therefore appears that the draft ISO partial flow procedure is appropriate for PM measurement under transient conditions and could be incorporated into the EU legislation.
The draft ISO standard also contains new procedures for the gravimetric measurement of PM based on studies undertaken to identify which elements of filter handling and weighing have a significant influence on repeatability and detection limit. It appears that all the draft ISO provisions regarding PM filter handling are a clear improvement on the existing procedures with the exception of repetition of test runs on the same filter.
The US EPA has recently adopted new procedures for PM measurement for the 2007 standards. New elements of the procedure include preconditioning of the dilution air, a narrow temperature range (42 to 52oC) for the filter holder, a particle pre-classifier to remove coarse, mechanically generated, particles as well as new handling and weighing techniques. The control of the filter holder temperature appears to be the most significant new requirement, because it controls the organic fraction of the PM, improving test repeatability. This is especially important for the measurement of PM from engines fitted with DPFs, as the soluble organic fraction (SOF) is the dominant component of the PM. It should be possible to adopt these improved techniques into the EC legislation.
The additional requirements of the US 2007 regulations increase the installation and operational cost of the full flow system by 10 to 15% for an existing installation. The cost of an up-to-date weighing room is about
Volume 2
Low Levels of Particulates
Page 5
EC-DG ENTR
2
Emission control technology for heavy-duty vehicles
ETD/00/503430
INTRODUCTION
Measuring Particulate Matter (PM) emitted from Euro 4 and 5 engines is a challenging task due to the very low levels that are likely to induce high uncertainties in the accuracy and repeatability of the existing sampling and measuring techniques. Difficulties arise mainly from the fact that the sampling procedure, as defined by the legislation, requires a substantial dilution of the exhaust flow, in order to decrease the sampling temperature below 52°C.1 The measuring procedure prescribed in the existing European legislation aims at ensuring that under laboratory conditions: 1. The engine operation is representative of real world operation. 2. The engine operation is repeatable. 3. The PM emission of the engine operating is representative of real world emissions. The latter two of the above are assumed to be achieved by the adoption and detailed description of specific engine operation cycles and sampling methodology (the extent to which these cycles represent real world operation of the engine is out of the scope of this study). However, the reduction by roughly one order of magnitude of the already low PM emission of current engines, foreseen in the near future, challenges the overall accuracy of the PM emission measurement as it is prescribed in the Directive 1999/96/EC. To understand the problems related with the measurement of low level PM requires first looking at the conventional measurement techniques used today. Most vehicle manufacturers and regulatory agencies utilise the Constant Volume Sampler (CVS). This exhaust dilution technique was applied systematically in automotive testing in the 1970’s as a tool for simulating atmospheric processes and minimising various collection problems. According to this technique, the total (or some fraction) of the exhaust stream is injected into a large duct and diluted with filtered ambient air. The sampling is conducted so that the total combined flow rate is nearly constant for all engine operating conditions. The basic aim is to permit sufficient flexibility to determine the size, physical and chemical nature of PM, and their relation to engine and sampling variables. This is done by appropriate design of the system to ensure sufficient mixing times for physical and chemical processes to occur. Although incremental improvements were made over the last 30 years, the technique has remained basically the same and its intrinsic problems impose difficulties in measuring emissions from newer vehicles. These problems concern the Dilution Ratios (DR) achieved that may be too high leading to concentrations below the systems’ detection limits, or too low for high speed and/or high acceleration driving schedules. Furthermore, low DR may be a problem for alternatively fuelled vehicles using fuels such as ethanol, methanol and compressed natural gas (CNG) whose combustion produces more water. In addition, a larger fraction of the PM mass is likely to be in the form of volatile matter, which is more subject to sampling errors than is solid carbonaceous material. 1
According to the existing legislation [1] (Annex I, clause 2.7), “Particulate pollutants means any material
collected on a specified filter medium after diluting the exhaust with clean filtered air so that the temperature does not exceed 325 K (52° C)”. Volume 2
Low Levels of Particulates
Page 7
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
In this context and according to the contract with the European Commission, this report attempts to describe in sufficient detail and provide costs of improved emission sampling and measurement techniques that can be used for the measurement of the very low levels of particulate mass emissions expected from high technology engines meeting the Euro 4 and Euro 5 emission limits. The accuracy and repeatability of such techniques is also assessed using published data. The study considered both diesel and gas fuelled heavy-duty vehicles.
Volume 2
Low Levels of Particulates
Page 8
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
2.1 APPROACH In order to fulfil the goals of the study, this report has the following structure, which reflects to a very large extent the way the issue was approached by the study team: •
In chapter 3 the main elements of the current and proposed legislation in Europe (Directive 1999/96/EC) and the US (2007 regulations for heavy-duty engines), as well as of the proposed ISO standard (ISO 16183) are briefly presented. The description focuses on the issues relevant to PM measurement only.
•
In chapter 4, after a summary of the fundamental considerations pertinent with the measurement of particles, a detailed comparison is made of the provisions of the 3 regulatory documents, presented in a tabular format. A discussion of the differences and similarities follows, supported and complemented with published data, as far as available. In parallel, published data are used in order to evaluate the capabilities of the currently available partial flow systems. The discussion is further supported with personal communications with vehicle manufacturers, instrument manufacturers and certification laboratories.
•
In the 5th chapter, the report looks at alternative measurement methods for the measurement of particulate mass, which could be of interest for legislative purposes in the near future. In this context it first looks at a new technique recently developed for the measurement of the PM mass via vaporisation at high temperatures, sulphate deoxidation, conversion of SOF and oxidation of carbon. As the ability to measure PM on a real time basis is as valuable to the engine research as it is for regulatory and environmental purposes, this chapter covers also the PM measurement possibilities in real time.
•
The costs associated with the various possibilities for both sampling and PM measurement systems are summarised and discussed in chapter 6.
•
Chapter 7 discusses the issue of measuring PM from gas (CNG) fuelled engines, using data published in the international literature.
•
The report concludes with conclusions and recommendations. An annex provides technical data on the currently available partial flow dilution systems.
Volume 2
Low Levels of Particulates
Page 9
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
3 THE EXISTING AND PROPOSED PM MASS MEASUREMENT TECHNIQUES This chapter briefly reviews the PM measurement provisions of the existing European legislation (Directive 1999/96/EC [1]), the US 2007 regulation ([2]) and of the proposed standard ISO 16183 [3].
3.1 THE DIRECTIVE 1999/96/EC 3.1.1 Dilution and Sampling System The European regulations for heavy-duty diesel engines are commonly referred to as Euro I ... V. Euro I standards for medium and heavy-duty engines were introduced in 1992. The Euro II regulations came to power in 1996. These standards applied to both heavy-duty highway diesel engines and urban buses. In 1999, the European Parliament and the Council of Environment Ministers adopted the final Euro III standard (Directive 1999/96EC of December 13, 1999, amending the Heavy Duty Diesel emissions Directive 88/77/EEC) and also adopted Euro IV and V standards for the years 2005 and 2008 respectively. The standards also set specific, stricter values for extra low emission vehicles (also known as "enhanced environmentally friendly vehicles" or EEVs) in view of their contribution to reducing atmospheric pollution in cities. In April 2001, the European Commission adopted Directive 2001/27/EC, which introduced further amendments to Directive 88/77/EEC. The new Directive prohibits the use of emission "defeat devices" and "irrational" emission control strategies, which would be reducing the efficiency of emission control systems when vehicles operate under normal driving conditions to levels below those determined during the emission testing procedure. Mass determination of the PM is currently carried out in accordance with the procedure prescribed by the Directive 1999/96/EC. Sampling is made according to the CVS (Constant Volume Sampler) technique. The idea behind this procedure is to dilute the vehicle’s entire raw exhaust with enough ambient air such that water condensation does not occur in the mixture (mixture temperature less than 52°C). A blower or pump draws the mixture through the system at a relatively constant combined volumetric flow rate. Thus, the system operates at a variable DR (as the vehicle produces more exhaust, less ambient air is mixed with it in order to keep the total flow constant). A measuring or metering device in the bulk stream determines the flow rate. In this manner, the total volume of the mixture is easy to determine by taking the time of the sampling and multiplying it by the constant flow rate. There are two types of CVS units accepted by the Directive: the Positive Displacement Pump (PDP) and the Critical Flow Venturi (CVF) types. The modern devices have taken advantage of electronics and computer technology to lower the cost and improve precision. Figure 1 illustrates the full dilution system (FDS) to obtain PM on the filters.
Volume 2
Low Levels of Particulates
Page 10
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Figure 1. The full flow dilution system according to the Directive 1999/96/EC. DAF = Dilution Air Filter, EP = Exhaust Pipe, PSP = Particulate Sampling Probe, PTT = Particulate Transfer Tube, HE = Heat Exchanger, EFC = Electronic Flow Compensation, PDP = Positive Displacement Pump, CVF = Critical Flow Venturi, FC = Flow Controller
A small proportional sample of the gas from the dilution tunnel is drawn through the filter to collect PM sample. If this is done directly, it is referred to as single dilution. If the sample is diluted once more in a secondary dilution tunnel, it is referred to as double dilution. The entire process involves the following steps: i.
Dilution of the full exhaust gas stream in the main tunnel duct ,
ii.
Monitoring of temperature in the dilution tunnel,
iii.
Weighing of filter, before and after the sampling in a chamber of controlled temperature and
humidity, using a microbalance, and iv.
Measurement of the flow rate of diluted exhaust that passes through the filter during the sample
period According to this method, a single PM value is available for the whole test and is expressed in units of mass per power time [g/kWh] or mass per brake horsepower-hour [g/bhph] for engine tests and mass per distance [g/km] for chassis tests. The option of double dilution is also considered in the Directive as a specific modification of a typical PM sampling system (Figure 2) and is useful if the filter temperature requirement cannot be met with single dilution. A schematic of the method is shown in Figure 3. It includes all the important parts of the PM sampling system (filter holders, sampling pump) and dilution features.
Volume 2
Low Levels of Particulates
Page 11
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Figure 2. Particulate Sampling System according to the Directive 1999/96/EC. FH = Filter Holder, P = Pump, EGA = Exhaust gas Analyser, GFUEL = Fuel mass Flow rate, FM = Flow Measurement (Device), BV = Ball Valve
Figure 3. Double Dilution system. SDT = Secondary Dilution Tunnel (Directive 1999/96/EC)
The recommendations for the FDS according to the Directive can be summarised as follows:
•
Maximum length 10 m (from the engine exhaust manifold, turbocharger outlet or aftertreatment device to the dilution tunnel)
•
Insulation of all tubing in excess of 4 m downstream of the above, except for an in-line smokemeter, if used. Thickness and insulating material properties are prescribed
•
Limited use of flexible sections (length to diameter ratio of 12 or less)
Volume 2
Low Levels of Particulates
Page 12
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Positive Displacement Pump, PDP •
The exhaust system backpressure must not be artificially lowered by the PDP or dilution air inlet system
•
Static exhaust backpressure measured with the PDP system operating shall remain within ±1,5 kPa of the static pressure measured without connection to the PDP at identical engine speed and load
•
Gas mixture temperature immediately ahead of the PDP shall be within ± 6 K of the average operating temperature observed during the test, when no flow compensation is used
•
The latter may only be used if the temperature at the inlet to the PDP does not exceed 50°C
Constant Flow Venturi, CFV •
Static exhaust backpressure measured with the CFV system operating shall remain within ±1,5 kPa of the static pressure measured without connection to the CVF at identical engine speed and load
•
Gas mixture temperature immediately ahead of the CFV shall be within ± 11 K of the average operating temperature observed during the test, when no flow compensation is used
Heat Exchanger, HE •
The heat exchanger shall be of sufficient capacity to maintain the temperature within the limits required above
Electronic flow compensation, EFC •
Electronic flow compensation is required if HE is not used to keep the temperature at the inlet of PDP or CFV within the above stated limits
Dilution Tunnel, DT •
Small enough in diameter to cause turbulent flow and sufficient length to cause complete mixing; a mixing orifice may be used
•
At least 460 mm in diameter with single system, at least 210 mm in diameter with a double system
•
May be insulated
•
Sufficient flow capacity to maintain the diluted exhaust at a temperature of less than or equal to 52°C immediately before the PM filter, when the single system is used
•
If secondary dilution system is used, sufficient flow capacity to maintain the diluted exhaust stream in the primary DT at a temperature of less than or equal to 191°C at the sampling zone is required. The secondary dilution system must provide sufficient air to maintain the doubly-diluted exhaust stream at a temperature of less than or equal to 52°C immediately before the PM filter
Volume 2
Low Levels of Particulates
Page 13
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Dilution Air Filter, DAF •
Filtration and charcoal scrubbing for background hydrocarbons elimination
•
At the engine manufacturers’ request the dilution air shall be sampled according to good engineering practice to determine the background PM levels, which can then be subtracted from the values measured in the diluted exhaust
Particulate Transfer Tube, PTT •
Not to exceed 1020 mm in length and minimised whenever possible. Dimension is valid from the tip of the probe to the filter holder for the single system and from the tip of the probe to the secondary dilution tunnel for the double method
•
Heating to be no greater than 52°C wall temperature by direct heating or by dilution air pre-heating, provided the air temperature does not exceed 52°C prior to the introduction of the exhaust in the DT
•
Insulation acceptable
Secondary Dilution Tunnel, SDT •
Minimum diameter of 75 mm with sufficient length to provide a residence time of at least 0,25 seconds for the doubly diluted sample
•
Location of the primary filter holder within 300 mm of the exit of SDT
•
Heating to no greater than 52°C wall temperature by direct heating or by dilution air preheating, provided the air temperature does not exceed 52°C prior to the introduction of the exhaust in the DT
•
Insulation acceptable
Filter Holder, FH •
The requirements specified in the subchapter titled Determination of Particulates shall be met (see 3.1.2)
•
Heating to no greater than 52°C wall temperature by direct heating or by dilution air pre-heating, provided the air temperature does not exceed 52°C prior to the introduction of the exhaust in the DT
•
Insulation acceptable
Pump, P •
Sampling pump shall be located sufficient) distant from the tunnel so that the inlet gas temperature is maintain at a constant temperature (±3 0C), if flow correction is not used
Dilution Air Pump, DP •
If secondary dilution is used the dilution air pump shall be located so that the secondary dilution air is supplied at a temperature of 25°C ± 5°C, if preheating is not used
Volume 2
Low Levels of Particulates
Page 14
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Flow Controller, FC •
A flow controller shall be used to compensate sample flow rate for temperature and pressure variations in the path, if no other means are available. Required if electronic flow compensation is used
Flow Measurement, FM •
The sample flow measurement device (FM3 of Figure 2 and Figure 3) shall be located sufficiently distant from the sampling pump so that the inlet gas temperature remains constant (±3 K), if flow correction by FC is not used
•
The flow measurement device (FM4 of Figure 3) for the dilution air shall be located so that the inlet gas temperature remains at 25°C ±5°C if preheating is not used.
Another type of sampling considered by the Directive is the Partial Flow Dilution System (PFDS), only in the case of steady state testing. In this method, only a portion of the exhaust gas is passed through the dilution tunnel. If the entire diluted exhaust gas is passed through the PM sampling system the method is referred to as total sampling type; if only a portion of the diluted exhaust gas is passed through the sampling system the method is referred to as fractional sampling type. There are various configurations that are described in detail in the Directive. Figure 4 illustrates one of the recommended configurations for the total sampling method and Figure 5 one of the configurations for the fractional method. Apart from the specific recommendations that concern each configuration, the general guidelines can be summarised as follows:
Figure 4. Partial flow dilution system with total sampling according to the Directive 1999/96/EC. Gi = Mass flow rate, SP = Sampling Probe, TT = Transfer Tube, DT = Dilution Tunnel
Volume 2
Low Levels of Particulates
Page 15
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Figure 5. Partial flow dilution system with flow control and fractional sampling (SB: Suction Blower) control according to the Directive 1999/96/EC. PB = Pressure Blower, ISP = Isokinetic Sampling Probe, PDT = Pressure Transducer, TT = Transfer Tube, PSP = Particulate Sampling Probe, PTT = Particulate Transfer Tube
Ball Valve, BV •
Optional; If used shall have an inside diameter not less than the inside diameter of the PM transfer tube, and a switching time of less than 0,5 seconds
Exhaust Pipe, EP •
It may be insulated
•
Thickness to diameter ratio of 0,015 or less to reduce thermal inertia
•
Length to diameter ratio of 12 or less
•
Minimisation of bends to reduce inertial deposition
•
Insulation of the test bed silencer if included
•
Pipe free of elbows, bends and sudden diameter changes for at least 6 pipe diameters upstream and 3 pipe diameters downstream of the tip of the probe to ensure isokinetic sampling
Transfer Tube, TT •
As short as possible. No more than 5 m in length
•
Equal to or greater than the probe diameter, but not more than 25 mm in diameter
•
Exiting on the centreline of the dilution tunnel and pointing downstream
•
Insulation with the appropriate material depending on the pipe length
Volume 2
Low Levels of Particulates
Page 16
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Dilution Tunnel, DT •
Shall be of a sufficient length to cause complete mixing of the exhaust and dilution air under turbulent flow conditions
•
Thickness/diameter ratio of 0,025 or less for DT with inside diameters greater than 75 mm
•
Nominal thickness of no less than 1,5 mm for the dilution tunnels with inside diameters of equal to or less than 75 mm
•
Shall be at least 75 mm in diameter for the fractional type
•
Shall be at least 25 mm in diameter for the total type
•
May be heated to no greater than 52°C wall temperature by direct heating or by dilution air preheating, provided that the air temperature does not exceed 52°C prior to the introduction of the exhaust in the dilution tunnel
•
Mixing orifice may be used to optimise mixing
•
May be insulated
•
If the dilution air in the vicinity of the DT is below 20°C, precautions (heating and/or insulation) should be taken to avoid particle losses onto the cold walls.
Particulate Sampling probe, PSP
The probe: • Shall be installed facing upstream at a point where the dilution air and exhaust gas are well mixed, i.e. on the dilution tunnel (DT) centreline approximately 10 tunnel diameters downstream of the point where the exhaust enters the dilution tunnel •
Shall be of 12 mm minimum inside diameter
•
May be heated to no greater than 325 K (52 °C) wall temperature by direct heating or by dilution air pre-heating, provided the air temperature does not exceed 325 K (52 °C) prior to the introduction of the exhaust in the dilution tunnel
•
may be insulated
3.1.2 Determination of Particulates The flow capacity of the dilution system shall be large enough to completely eliminate water condensation in the dilution and sampling systems, and maintain the temperature of the diluted exhaust gas at or below 52°C immediately upstream of the filter holders. Dehumidifying the dilution air before entering the dilution system is permitted, and is especially useful if the dilution air humidity is high. The temperature of the dilution air shall be 25°C ±5°C. If the ambient temperature is below 20°C, dilution air pre-heating above the upper temperature limit of 30°C is recommended. However, the dilution air temperature must not exceed 52°C prior to the introduction of the exhaust in the dilution tunnel.
Volume 2
Low Levels of Particulates
Page 17
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
To determine the mass of the particulates, a particulate sampling system, particulate sampling filters, a microgram balance, and a temperature and humidity controlled weighing chamber, are required. For particulate sampling, the single filter method shall be applied which uses one pair of filters for the whole test cycle. For the ESC, considerable attention must be paid to sampling times and flows during the sampling phase of the test.
Particulate Sampling Filters •
Filter Specification. Fluorocarbon coated glass fibre filters or fluorocarbon based membrane filters are required. All filter types shall have a 0,3um DOP (di-octylphthalate) collection efficiency of at least 95% at a gas face velocity between 35 and 80 cm/s.
•
Filter Size. Particulate filters must have a minimum diameter of 47 mm (37 mm stain diameter). Larger diameter filters are acceptable.
•
Primary and Back-up Filters. The diluted exhaust shall be sampled by a pair of filters placed in series (one primary and one back-up filter) during the test sequence. The back-up filter shall be located no more than 100 mm downstream of, and shall not be in contact with the primary filter. The filters may be weighed separately or as a pair with the filters placed stain side to stain side.
•
Filter Face Velocity. A gas face velocity through the filter of 35 to 80 cm/s shall be achieved. The pressure drop increase between the beginning and the end of the test shall be no more than 25 kPa.
•
Filter Loading. The recommended minimum filter loading shall be 0,5 mg/1075 mm2 stain area. For the most common filter sizes the values are shown in Table 1.
Table 1. Recommended filter loading according to Directive 1999/96/EC
Filter Diameter
Recommended stain
Recommended
(mm)
(mm)
Minimum Loading (mg)
47
37
0,5
70
60
1,3
90
80
2,3
110
100
3,6
Weighing Chamber and Analytical Balance Specifications •
Weighing Chamber Conditions. The temperature of the chamber (or room) in which the particulate filters are conditioned and weighed shall be maintained to within 22 ± 3°C during filter conditioning and weighing. The humidity shall be maintained to a dew point of 9,5 ± 3°C and a relative humidity of 45±8%.
•
Reference Filter Weighing. The chamber (or room) environment shall be free of any ambient contaminants (such as dust) that would settle on the particulate filters during their stabilisation. Disturbances to weighing room specifications as outlined above will be allowed if the duration of the distur-
Volume 2
Low Levels of Particulates
Page 18
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
bances does not exceed 30 minutes. The weighing room should meet the required specifications prior to personal entrance into the weighing room. At least two unused reference filters or reference filter pairs shall be weighed within 4 hours of, but preferably at the same time as the sample filter (pair) weightings. They shall be the same size and material as the sample filters. •
If the average weight of the reference filters (reference filter pairs) changes between sample filter weightings by more than ±5 % (±7,5% for the filter pair respectively) of the recommended minimum filter loading then all sample filters shall be discarded and the emissions test repeated
•
If the weighing room stability criteria is not met, but the reference filter (pair) weightings meet the above criteria, the engine manufacturer has the option of accepting the sample filter weights or voiding the tests, fixing the weighing room control system and rerunning the test.
•
Analytical Balance. The analytical balance used to determine the weights of all filters shall have a precision (standard
! !
mm diameter, the !
=10 " #
Additional Specifications for Particulate Measurement
All parts of the dilution system and the sampling system from the exhaust pipe up to the filter holder, which are in contact with raw and diluted exhaust gas, must be designed to minimise deposition or alteration of the particulates. All parts must be made of electrically conductive materials that do not react with exhaust gas components, and must be electrically grounded to prevent electrostatic effects.
Volume 2
Low Levels of Particulates
Page 19
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
3.2 THE DRAFT ISO FDIS 16183 (VERSION 195E) This draft standard specifies the measurement and evaluation methods for gaseous and particulate exhaust emission from heavy-duty engines under transient conditions on a test bed. The procedure defined can be applied to any transient test cycle that does not require extreme system response times. Therefore, it may be used as an option to the regulated measurement equipment of certification test cycles (usually CVS type systems) with the approval of the certification agency. Directive 2001/27/EC has already adopted the technical prescriptions of ISO FDIS 16183 (dated 15 October 2000) for NOx emissions measurements during the ETC screening test, as an alternative to the requirements of Directive 88/77/EEC. With respect to particulate measurement, the use of the partial dilution system is currently being examined in detail. The complete measurement set-up (including gas emissions) is schematically shown in Figure 6.
Figure 6. Schematic of the measurement set-up of a partial flow dilution system according to [2]
3.2.1 Dilution and Sampling System The dilution configurations and sampling system recommended in this standard are those prescribed by the Directive and shown in Figure 4, Figure 5 and Figure 2. Two innovations are adopted: the first one concerns the test procedure, and the second concerns the dilution air temperature. Specifically, measurements under engine transient conditions are prescribed and dilution air temperature higher than 15°C is recommended.
Volume 2
Low Levels of Particulates
Page 20
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
3.2.2 Determination of Particulates
Particulate Sampling Filters •
Filter Specification. Innovation concerns the specified collection efficiency (at least 99% at a gas face velocity between 35-100 cm/s).
•
Filter size. As in Directive 1999/96/EC.
•
Filter Face Velocity. Increase of the upper limit to 100 cm/s
•
Filter Loading. The required minimum filter loading for the most common filter sizes are shown in Table 2. For larger filter sizes, the minimum filter loading shall be 0,065 mg/1000 mm² filter area
Table 2. Minimum filter loading according to ISO 16183 Filter Diameter (mm) Minimum loading (mg)
47
0,11
70
0,25
90
0,41
110
0,62
Weighing Chamber and Analytical Balance Specifications •
Weighing Chamber Conditions. As in Directive 1999/96/EC.
•
Reference Filter Weighing. As in Directive 1999/96/EC.
•
Analytical Balance. The analytical balance used to determine the filter weight shall have a precision (standard deviation) of at least 2 µg and a resolution of at least 1 µg (1 digit = 1 µg) specified by the $ % !
resolution balance is required. If unstable or irreproducible filter weightings are observed due to the effects of static electricity, the filters shall be neutralised prior to weighing, e.g. by a Polonium neutraliser or a device of similar effect.
Volume 2
Low Levels of Particulates
Page 21
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
3.3 US 2007 FEDERAL REGULATIONS US EPA proposed a particulate matter (PM) emission standard for new heavy-duty engines of 0.01 g/bhp-hr, to take full effect in the 2007 heavy-duty engine (HDE) model year. The adopted 2007 emission standards represent a landmark regulation of unprecedented stringency. Unlike cars, which commonly use catalytic converters, most of current heavy-duty engines rely on engine technologies to meet emission regulations and have not been using emission aftertreatment devices. The USA has followed the lead of the European Union in adopting emission standards that are designed to force emission aftertreatment devices on heavy duty diesel engines. The current EPA 2007 rule, however, sets emission limits that are much tighter than Euro IV and V, becoming the most stringent diesel emission standard worldwide. For comparison, today's PM standard is 0.1 g/bhp-hr for truck engines and 0.05 g/bhp-hr for urban buses.
3.3.1 Dilution and Sampling System The system prescribed utilises the CVS concept of measuring the combined mass emissions. Multiple or redundant systems may be used during a single test and statistical averages of data may be used to calculate results, consistent with good engineering judgement. The prescribed measurement set-up is illustrated in Figure 7. The components necessary for exhaust sampling shall meet the following requirements:
The flow capacity of the CVS must be sufficient to maintain the diluted exhaust stream at the temperatures required for the measurement of particulate emission noted below and at, or above, the temperatures where aqueous condensation in the exhaust gases could occur. This is achieved by the following method. The flow capacity of the CVS must be sufficient to maintain the diluted exhaust stream in the primary dilution tunnel at a temperature of 191°C or less at the sampling zone and as required to prevent condensation at any point in the dilution tunnel. Gaseous emission samples may be taken directly from this sampling point. An exhaust sample must then be taken at this point to be diluted a second time for use in determining particulate emissions. The secondary dilution system must provide sufficient secondary dilution air to maintain the double-diluted exhaust stream at a temperature of 47°C± 5°C, measured at a point located between the filter face and 16 cm upstream of the filter face.
The CVS system may use a PDP or a CFV. PDP systems must use a heat exchanger, while the CFV systems may use either a heat exchanger or electronic flow compensation. When electronic flow compensation is used, the CFV may be replaced by a subsonic venturi (SSV) as long as the CVS concept is maintained (i.e., a constant volumetric flow-rate through the CVS is maintained for the duration of the test).
When a heat exchanger is used, the gas mixture temperature, measured at a point immediately ahead of the critical flow venturi, shall be within 11°C of the average operating temperature observed during the test with the simultaneous requirement that aqueous condensation does not occur. The temperature measuring system (sensors and readout) shall have an accuracy and precision of 1.9°C. For systems util-
Volume 2
Low Levels of Particulates
Page 22
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
ising a flow compensator to maintain proportional sampling, the requirement for maintaining constant temperature is not necessary.
Figure 7. Schematic of the measurement set-up according to US EPA 2007 regulations
Volume 2
Low Levels of Particulates
Page 23
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
The primary dilution air and secondary dilution air: i.
Shall have a primary and secondary dilution air temperature equal to or greater than 15°C.
ii.
Primary dilution air shall be filtered at the dilution air inlet. The manufacturer of the primary di-
lution air filter shall state that the filter design has successfully achieved a minimum particle removal efficiency of 98% (less than 0,02 penetration) as determined using ASTM test method F 1471-93. Secondary dilution air shall be filtered at the dilution air inlet using a high-efficiency particulate air filter (HEPA). The HEPA filter manufacturer shall state the HEPA filter design has successfully achieved a minimum particle removal efficiency of 99,97% (less than 0,0003 penetration). It is recommended that the primary dilution air be filtered using a HEPA filter. EPA intends to utilise HEPA filters to condition primary dilution air in its test facilities. It is acceptable to use a booster blower upstream or downstream of a HEPA filter in the primary dilution tunnel (and upstream of the introduction of engine exhaust into the CVS) to compensate for the additional pressure loss associated with the filter. The design of any booster blower located downstream of the filter should minimise the introduction of additional particulate matter into the CVS. iii.
Primary dilution air may be sampled to determine background particulate levels, which can then
be subtracted from the values measured in the diluted exhaust stream. In the case of primary dilution air, the background particulate filter sample shall be taken immediately downstream of the dilution air filter and upstream of the engine exhaust flow.
The primary dilution tunnel shall be: i.
Small enough in diameter to cause turbulent flow (Reynolds Number greater than 4000) and of
sufficient length to cause complete mixing of the exhaust and dilution air. Good engineering judgement shall dictate the use of mixing plates and mixing orifices to ensure a well-mixed sample. To verify mixing, EPA recommends flowing a tracer gas (i.e. propane or CO2) from the raw exhaust inlet of the dilution tunnel and measuring its concentration at several points along the axial plane at the sample probe. Tracer gas concentrations should remain nearly constant (i.e. within 2%) between all of these points. ii.
At least 20 cm in diameter.
iii.
Constructed of electrically conductive material which does not react with the exhaust compo-
nents iv.
Electrically grounded.
v.
Have minimal thermal capacitance such that the temperature of the walls tracks with the tem-
perature of the diluted exhaust.
The temperature of the diluted exhaust stream inside of the primary dilution tunnel shall be sufficient to prevent water condensation.
Volume 2
Low Levels of Particulates
Page 24
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
The engine exhaust shall be directed downstream at the point where it is introduced into the primary dilution tunnel. The sampling system shall collect a proportional sample from the primary tunnel, and then transfer this sample to a secondary dilution tunnel where the sample is further diluted. The double-diluted sample is then passed through the collection filter. Proportionality (i.e., mass flow ratio) between the primary tunnel flow rate and the sample flow rate must be maintained within 5%, excluding the first 10 seconds of the test at start-up. The requirements of this system are: i.
The particulate sample transfer tube shall be configured and installed so that: a) The inlet faces upstream in the primary dilution tunnel at a point where the primary dilution air and exhaust are well mixed. b) The particulate sample exits on the centreline of the secondary tunnel.
ii.
The entire particulate sample transfer tube shall be: a) Sufficiently distant (radially) from other sampling probes (in the primary dilution tunnel) so as to be free from the influence of any wakes or eddies produced by the other probes. b) 0,85 cm minimum inside diameter. c) No longer than in 91 cm from inlet plane to exit plane. d) Designed to minimise the diffusional and thermophoretic deposition of particulate matter during transfer (i.e., sample residence time in the transfer tube should be as short as possible, temperature gradients between the flow stream and the transfer tube wall should be minimised). Double-wall, thin-wall, air-gap insulated, or a controlled heated construction for the transfer tube is recommended. e) Constructed such that the surfaces exposed to the sample shall be an electrically conductive material, which does not react with the exhaust components, and this surface shall be electrically grounded so as to minimise electrostatic particulate matter deposition.
iii.
The secondary dilution air shall be at a temperature equal to or greater than 15°C.
iv.
The secondary-dilution tunnel shall be constructed such that the surfaces exposed to the sample
shall be an electrically conductive material, which does not react with the exhaust components, and this surface shall be electrically grounded so as to minimise electrostatic particulate deposition. v.
Additional dilution air must be provided so as to maintain a sample temperature of 47° C±5° C
upstream of the sample filter. Temperature shall be measured with a thermocouple with a
shank,
having thermocouple wires with a wire size of 24 AWG (0.205 mm2) or smaller, a bare-wire butt-welded junction; or other suitable temperature measurement with an equivalent or faster time constant and an accuracy and precision of 1,9°C. vi.
The filter holder assembly shall be located within 30.5 cm of the exit of the secondary dilution
tunnel. vii.
Volume 2
The face velocity through the sample filter shall not exceed 100 cm/s.
Low Levels of Particulates
Page 25
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
3.3.2 Determination of Particulates
Particulate Sampling Filters Filter Specification. Polytetrafluoroethylene (PTFE or teflon) coated borosilicate glass fibre highefficiency filters or PTFE high-efficiency membrane filters with an integral support ring of polymethylpentene (PMP) or equivalent inert material are required. Filters shall have a minimum clean filter efficiency of 99% as measured by the ASTM D2986-95a DOP test. Filter Size. Must have a diameter of 46±0,6 mm (38 mm minimum stain diameter). A single highefficiency filter simultaneously samples the dilute exhaust during the cold-start test and a second high efficiency filter during the hot-start test. It is recommended that the filter loading should be maximised consistently with temperature requirements
Filter Holder Assembly. The filter holder assembly shall comply with the specifications set forth for ambient PM measurement in 40 CFR Part 50, Appendix L 7.3.5 with the following exceptions: i.
The material shall be 302, 303, or 304 stainless steel instead of anodised aluminium.
ii.
The 2.84 cm diameter entrance to the filter holder may be adapted, using sound engineering
judgement and leak-free construction, to an inside diameter no smaller than 0.85 cm, maintaining the 12.5 degrees angle from the inlet of the top filter holder to the area near the sealing surface of the top of the filter cartridge assembly. Similar variations using sound engineering design are also acceptable provided that they provide even flow distribution across the filter media and a similar leak-free seal with the filter cartridge assembly. iii.
If additional or multiple filter cartridges are stored in a particulate sampler as part of an auto-
matic sequential sampling capability, all such filter cartridges, unless they are installed in the sample flow (with or without flow established) shall be covered or sealed to prevent communication of semivolatile matter from filter to filter; contamination of the filters before and after sampling; or loss of volatile or semi-volatile particulate matter after sampling.
Particle Preclassifier. A particle preclassifier shall be installed immediately upstream of the filter holder assembly. The purpose of the preclassifier is to remove coarse, mechanically generated particles (e.g., rust from the engine exhaust system or carbon sheared from the sampling system walls) from the sample flow stream while allowing combustion-generated particles to pass through to the filter. The preclassifier may be either an inertial impactor or a cyclonic separator. The preclassifier manufacturer’s 50% cutpoint & !
PM emissions. Sharpness of cut is not specifically defined, but the preclassifier geometry shall allow at least 99% of the mass concentration of 1 m particles to pass through the exit of the preclassifier to the filter at the volumetric flow rate selected for sampling particulate matter emissions. Periodic servicing of the preclassifier will be necessary to prevent a build-up of mechanically separated particles. The particle preclassifier may be made integral with the top of the filter holder assembly. The preclassifier may also
Volume 2
Low Levels of Particulates
Page 26
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
be made integral with a mixing-tee for introduction of secondary dilution air, thus replacing the secondary dilution tunnel; provided that the preclassifier provides sufficient mixing.
Volume 2
Low Levels of Particulates
Page 27
EC-DG ENTR
4 DISCUSSION TECHNIQUES
Emission control technology for heavy-duty vehicles
OF
THE
PROVISIONS
OF
ETD/00/503430
THE
EXISTING
Table 3 summarises the most important provisions of 1999/96/EC, ISO 16183 and US Federal Register (FTP 2007) regarding the measurement of PM and allows direct comparisons. These comparisons are discussed in detail in the following sections. However, prior to this discussion, it was felt necessary to summarise some fundamental aspects concerning the measurement of particulates, on the basis of the relevant literature.
4.1 FUNDAMENTAL CONSIDERATIONS The following methods used to sample and dilute the exhaust may influence the measurement results:
The means of transporting the exhaust to the dilution tunnel,
The design of the tunnel,
The tunnel operating conditions: collection temperature, dilution ratio (DR), humidity
From the above, the importance of temperature is evident. All the physical processes (condensation and adsorption/desorption, outgassing, thermophoresis) that influence the measurement variability are temperature dependent and results in uncertainties that have been experienced by all laboratories. Mixture temperature is a function of the exhaust temperature, dilution air temperature and dilution ratio. Due to limitations in the DR value imposed by the maximum permitted mixture temperature on one hand, and by difficulties in collecting enough mass for accurate measurements on the other, there are significant sources of variability in the measurements (for example the use of highly non-adiabatic dilution causes thermophoretic phenomena that lead to PM transport to sampling system surfaces). Other uncertainties related to DR concern condensation and adsorption/desporption phenomena and hence the soluble organic fraction (SOF) of PM in general. Kittelson and Johnson [4] have studied in significant detail the sources of variability in PM measurements of heavy-duty engines. Their recommendations are listed below (in order of their importance and, to a lesser extent, qualitative cost effectiveness):
Reduction of thermophoretic deposition •
By preheating the tailpipe walls
•
By introducing dilution air into the tailpipe at the engine outlet
•
By using a thin wall, low thermal mass tailpipe
•
By making the tailpipe shorter
Engine and tunnel conditioning
Temperature and/or dilution ratio control
Volume 2
Low Levels of Particulates
Page 28
EC-DG ENTR
Emission control technology for heavy-duty vehicles
Improved flow measurements
Improvement of cycle matching
Emission test fuel
Improved filters and filter handling procedures
Conditioning of combustion and tunnel air, and
Deposition analysis of test facilities
ETD/00/503430
Many of the above recommendations are already tackled by the existing and future legislative test procedures. However, since the exhaust and the sampling system as well as some of the processes taking place are poorly understood, fundamental study is required to get optimum engineering solutions. For example, fundamental studies on deposition and re-entrainment would not only improve the ability to design repeatable particle sampling-measurement systems, but would also give an insight into the phenomena that influence the nature and quantity of PM emitted. The influence of adsorption/desorption and condensation/evaporation processes is also poorly understood. Even relatively basic information is not available. For example, the time constants for mass transfer of volatile matter to and from the surface of soot particles immersed in flowing gases are unknown. This determines how easily volatile material on the surface of particles will be lost or gained, and is an important factor in the sensitivity of the SOF to dilution ratio. The importance of these processes will grow as SOF increases. Thus for very low emission engines where a lubrication oil related SOF constitutes a major fraction of the particulate matter emitted, high sensitivity would be expected. Another factor, which may be important for high SOF engines, is the interaction of the volatile compounds with the surfaces of the collection filters. In addition, there is considerable disagreement on how much the mixture temperature at the filter face varies during the transient test. To confirm experimentally that the recommendations deduced from the fundamental studies will indeed make the desired improvements in repeatability, and will not significantly change the absolute level of measured PM emissions, it will be necessary to set up one or more test cells that incorporate some or all of these recommendations. Design of experimental studies should be based on the fundamental studies’ key variables. Additional investigations to help explain the difference between observed and predicted phenomena might include size distribution analysis and chemical analysis of the filter samples.
Volume 2
Low Levels of Particulates
Page 29
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Table 3. Comparison of the most important provisions of 1999/96/EC, ISO 16183 and US 2007 federal regulation
Important provisions of Directive 1999/96/EC, ISO 16183 proposal and US FTP 2007. DIRECTIVE 1999/96/EC
ISO 16183
US Federal Regulation
SAMPLING SYSTEM CONFIGURATION ESC: raw exhaust for gaseous
Transient tests with raw exhaust
Transient and steady-state tests with full
pollutants, partial dilution for PM, for gaseous pollutants and partial
dilution for gaseous emissions (maximum
optionally full dilution for both
sampling temperature 191°C). Secondary
dilution for PM
gaseous and PM, with 191°C
dilution for particulate emissions (for sam-
maximum sample temperature for
pling temperature see below)
gaseous and secondary dilution for PM ETC: full dilution for gaseous and PM (with 191°C maximum sample temperature for gaseous and secondary dilution for PM), partial dilution acceptable with the approval of Technical Service Dilution air: temperature
Dilution air: temperature
25±5°C, dehumidification per-
>=15°C, dehumidification per-
mitted, unless dilution air pre-
mitted
Primary and secondary dilution air temperature >=15°C. Air for primary dilution filtered with minimum filtration efficiency 98%.
heating is used
Air for secondary dilution filtered with HEPA with 99.97% minimum efficiency, recommended also for primary dilution. Background particulate load (primary dilution air) subtraction permitted. Diluted exhaust temperature at filter holder: maximum 52°C
Diluted exhaust temperature at filter holder: maximum 52°C
Diluted exhaust temperature at filter holder: 47±5°C. Sampling at 30.5 cm (12’’) maximum from secondary dilution tunnel exit. A particle pre-classifier must be installed immediately upstream the filter holder, either inertia impactor or cyclonic separator, with 50% cut point between 2.5 and 10
$ ''( !
particles to pass through, in terms of mass
Volume 2
Low Levels of Particulates
Page 30
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Important provisions of Directive 1999/96/EC, ISO 16183 proposal and US FTP 2007. DIRECTIVE 1999/96/EC
ISO 16183
US Federal Regulation
SAMPLE HANDLING Minimise deposition or alteration Minimise deposition or alteration Minimise diffusional and thermophoretic of particulates, all system parts of particulates, all system parts deposition during transfer. Double wall, electrically
conductive
and electrically
conductive
and thin wall, air gap insulated or controlled
grounded to prevent electrostat- grounded to prevent electrostatics heated sample piping recommended. ics effects, dilution tunnel and effects, dilution tunnel and sam- Piping surfaces in contact with the sample sample piping walls may be ple piping walls may be heated must be electrically conductive and heated up to 52°C
up to 52°C.
grounded.
PM SAMPLING – FILTERS Fibre or PTFE filter, minimum PTFE filter, minimum efficiency PTFE or TEFLON coated glass fibber effi ) *+,- '&(
) *+,- ''(
filters, minimum efficiency > 99%.
Filter face velocity: 35 – 80 cm/s
Filter face velocity: 35 – 100 Maximum filter face velocity 100 cm/s cm/s
Filter size: min. diameter 47mm, Filter size: min. diameter 47 mm, Minimum filter diameter: 46.5±0.6 mm acceptable also 70, 90, 110 mm
acceptable also 70, 90, 110 mm
Stain diameter defined as (filter Stain diameter not defined
Minimum stain diameter 38 mm
diameter – 10 mm) FILTER LOADING Minimum recommended (at the Minimum required (but lower No specific values are given. It is recommanufacturers request, the ESC also allowed, with the agreement mended to maximise filter loading test sequence may be repeated a of the parties involved and with sufficient number of times for the use of a balance of higher sampling more PM mass on the resolution. Alternatively, a secfilter)
ond run of the test cycle on the same sample filters is permitted, with the agreement of the parties involved)
47 mm ⇐ 0,5 mg
47 mm ⇐ 0,11 mg
70 mm ⇐ 1,3 mg
70 mm ⇐ 0,25 mg
90 mm ⇐ 2,3 mg
90 mm ⇐ 0,41 mg
110 mm ⇐ 3,6 mg
110 mm ⇐ 0,62 mg
(resulting for the general recom- (resulting from the general remendation 0,5 mg / 1075 mm2 quirement of 0,065 mg / 1000 stain area)
Volume 2
mm2 filter area)
Low Levels of Particulates
Page 31
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Important provisions of Directive 1999/96/EC, ISO 16183 proposal and US FTP 2007. DIRECTIVE 1999/96/EC
ISO 16183
US Federal Regulation
FILTER BACKPRESSURE INCREASE DURING SAMPLING Maximum 250 mbar (25 kPa)
Maximum 250 mbar (25 kPa)
Not defined
MASS FLOW RATE OF THE SAMPLE The ratio of dilution tunnel –
The ratio of dilution tunnel – sample flow
sample flow rates must be main-
rates must be maintained within ±5%.
tained within ±5% of the set value. NUMBER OF FILTERS 2 (primary and backup)
1
1
FILTER CONDITIONING (empty and loaded) Temperature: 22°C ± 3°C Relative
humidity:
Temperature: 22°C ± 3°C
Temperature: 22°C ± 3°C
± Relative humidity: 45% ± 8% Dew point 9.5 ± 1°C
45%
(dew point: 9.5 ± 3oC)
8%(dew point: 9.5 ± 3oC) SAMPLE FILTER WEIGHING
Room conditions: as for condi- Room conditions: as for condi- Room conditions: as for conditioning tioning
tioning
Analytical balance: resolution 1 Analytical balance: resolution 1 Analytical balance: precision (standard
.
. & !
e- sizes. If filter loading lower than
# !
/0#
mm)
the minimum required is acl-
!
ance is required. Reference filter weighing permissible deviation: 5% of the
Reference filter weighing permissible deviation: maxi !
recommended minimum filter
Reference filter weighing permissible - !
filters are recommended)
loading The weighing results must be corrected for buoyancy effects. Specific calculation scheme is given, based on ambient conditions (barometric pressure, temperature, humidity) and the densities of sample filter and calibration weight
Volume 2
Low Levels of Particulates
Page 32
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Important provisions of Directive 1999/96/EC, ISO 16183 proposal and US FTP 2007. DIRECTIVE 1999/96/EC
ISO 16183
US Federal Regulation
ADDITIONAL SPECIFICATIONS In case of unstable balance reading, the use of neutraliser is rec-
Static electricity neutraliser is generally recommended.
ommended, to eliminate electrostatic effects
4.2 SAMPLING SYSTEM CONFIGURATIONS As regards dilution and sampling, the current 1999/96/EC Directive has in principle adopted a number of possible configurations. For steady state cycles these include raw exhaust sampling for gaseous pollutants, partial flow dilution for PM and full flow dilution for both gaseous and PM emissions. For transient cycles they include full flow dilution for both gaseous and PM emissions, accepting in parallel the partial dilution with the approval of Technical Service. Still the majority of (if not all) type approvals granted on the basis of this directive is based on measurements with full flow CVS, as no detailed specifications have existed for partial flow systems and raw exhaust measurements. As already mentioned, Directive 2001/27/EC has only adopted the technical prescriptions of ISO FDIS 16183 (dated 15 October 2000) for NOx emissions measurements during the ETC screening test, as an alternative to the requirements of Directive 88/77/EEC. On the other hand, US EPA in January 2001 adopted new regulations for PM measurement for the 2007 emission standards. These regulations concentrate on the full flow CVS technique without adopting the partial flow systems. Full flow dilution has been in question for some time, since it needs large CVS systems to achieve adequate dilution ratios for the large heavy duty engines. Capacities of the order of 180 m3/min are typical for engines up to 500 kW, in order to achieve dilution ratios of the order of 10 at the nominal mode of operation. These high capacities are associated with relatively large dilution ratios at partial loads, which in the low emitting engines of the near future lead to extremely low concentrations of gaseous species and particulates, which may severely affect the accuracy and repeatability of the test results. On the other hand the installation and operating costs of these systems are very high, especially for large engines. Moreover, the need for thorough dilution air preconditioning, including dehumidification, ambient air particle filtration and scrubbing and/or catalytic conditioning leads to expensive installations for these large volumes. Typical purchase cost of a full flow dilution system of a capacity of 180 m3/min is in the range of 600 k, including the CVS, the dilution tunnel, the secondary dilution and ancillaries. Partial flow dilution systems attempt to avoid these limitations by reversing the order of diluting and sampling of the exhaust gases, as a small sample of the exhaust gas is first taken and then accurately diluted at a
Volume 2
Low Levels of Particulates
Page 33
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
constant dilution ratio. The type of device used is that labelled by the industry as a “Mini-diluter” [5, 6, 7, 8]. The theory of operation is described in detail in [9], which lists its advantages as follows:
Less dilution is used because the diluent is dry, resulting in higher concentrations,
The diluent contains zero or predictable low levels background contaminants, so no background measurement is needed,
The water vapour compensation technique (water vapour sensor and correction factor to eliminate the inaccuracies caused by lost water vapour) compensates for changes in sample gas composition during cold start phases,
Internal HC hang-up is reduced by the simpler sampler.
Objective of the new ISO standard (ISO/FDIS 16183 - still under voting) was to develop the necessary specifications for the measurement of gaseous emissions from raw exhaust gas and of particulate emissions using partial flow dilution systems under transient test conditions. In the framework of UN-ECE GRPE activities, a number of correlation studies were undertaken by the Worldwide Heavy Duty Certification (WHDC) group to unravel the relationship between the conventional CVS technique and partial flow dilution. In addition, these studies have looked at a number of details of the conventional gravimetric measurement techniques, introduced by the ISO/FDIS 16183 [10, 11, 12]. The terms of reference of the WHDC working group, with respect to PM, were [10, 13]:
Analysis of current and alternative measurement procedures, Accuracy of current measurement procedures as regards future low emitting engines, Analysis of flow compensation systems for transient engine operation, Development of calculation procedure, Correlation study, and Correlation between partial and full flow systems for PM emission determination.
The tests were conducted in parallel in a number of laboratories (incl. EMPA in Switzerland, RWTÜV in Germany, JARI in Japan and SwRI in the US) and a number of available mini-dilution systems were used for the tests (AVL’s Smart Sampler, Horiba’s Partial Flow Dilution Tunnel, Pierburg’s Partial Dilution System, Control System’s PSS-20, Microtrol’s Transient Minitunnel and Sierra’s BG-2). Annex I presents the technical data for these mini dilutors.
Volume 2
Low Levels of Particulates
Page 34
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
4.2.1 Transient control capability of partial flow dilution systems
Figure 8 Transient sampling during the ETC with Control System PSS-20 (G_exhaust: exhaust gas mass flow, G_samp: sample gas flow, G_samp-0,2s: sample gas flow with delay of 0,2s) [10]
The fundamental requirement of a partial flow dilution system used over transient tests is its ability to proportionally sample. The latest developments in fast response flow rate control can guarantee that the exhaust volume that enters the tunnel is proportional to the total engine exhaust flow rate. The transient control capability of partial flow dilution systems was proven in all the correlation studies, via regression analysis. As an example, Figure 8 shows the comparison between the exhaust flow rate and the sample flow rate for the PSS 20, during a portion of the ETC. It should be stressed at this point that the latest ISO 16183 Version 195E tightened the requirements for response time of a partial flow dilution system to ≤ 0,3 seconds, compared to ≤ 0,5 seconds of the previous version 182E.
4.2.2 Essential parameters that may influence PM measurement with partial flow systems The PM correlation studies investigated a number of important parameters that can influence PM mass and composition. These were:
Parameters essential for both full flow and partial dilution systems: Dilution ratio, Filter face velocity, Sample filter loading.
Parameters essential for partial systems only: Sample line temperature, Sample line diameter, Sample line length, Sample probe design.
Table 4 presents the summary of the results of statistical analysis (Analysis of Variance) performed with the data from EMPA [10]. Statistically, the most significant parameter was the dilution ratio, as expected. From Volume 2
Low Levels of Particulates
Page 35
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
the other general parameters, filter face velocity and filter loading showed only very few significant effects. PM levels tended to be slightly higher at low filter face velocity and low filter loading. This finding would allow lower minimum filter loading in future emissions regulations to take account of the low PM levels of those engines. For the parameters related to partial flow dilution systems, significant effects were only observed in a few cases, and they were not consistent. PM levels tended to be slightly lower with a longer sample line so sample lines shorter than 1.5 m are recommended. For sample line temperature, tunnel heating and sample line diameter the current legislative requirements seem to be acceptable.
A similar statistical analysis of the experimental data of RWTÜV indicated that
There was no statistically significant difference between the four sample probe designs tested - although the open probe design was closer to the CVS.
The influence of the dilution air temperature and the sample line temperature was only minor,
Dilution ratio had a slight influence in most cases as illustrated in Figure 9. Nevertheless, the differences in absolute numbers were very small (between 0.002 and 0.005 g/kWh). ESC and ETC results are more consistent with a higher PM at higher dilution ratio under all dilution air and sample line temperature combinations.
The detailed results of similar correlation studies conducted in Japan and the US are not available.
Volume 2
Low Levels of Particulates
Page 36
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Table 4. Analysis of variance (ANOVA) results of parameter study at EMPA [10]
) #&
' # (# $
' #
%# %&
&# " %# # # " %# # ! ! " "#$
Figure 9. Influence of dilution ratio on PM emission (B25 and B100) based on RWTÜV data [10] DAT = Dilution Air Temperature, SLT = Sampling Line Temperature.
Volume 2
Low Levels of Particulates
Page 37
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
4.2.3 Correlation of partial flow systems with CVS With regard to PM emissions, overall, the results of the correlation studies published so far [10, 11, 12, 14, 15, 16] indicate the following:
According to [10], partial flow dilution systems measured slightly lower (2% to 15% on average) PM than CVS systems on steady state and transient cycles. However, there was a scatter between the different laboratories: •
At EMPA, the difference was basically statistically non significant (partial flow systems tested: Smart Sampler SPC 472 provided by AVL and PSS-20 by Control System);
•
At JARI and RWTÜV, the differences were greater and mostly statistically significant (partial flow systems tested: Horiba MDLT-1300T at JARI and Smart Sampler SPC 472 together with the Transient Minitunnel MICROTROL NOVA system at RWTÜV);
•
At Southwest Research Institute (SwRI), the correlation was very poor compared to the above studies and to current knowledge (partial flow system tested: BG-2 by Sierra).
In contrast to the above, additional tests reported in [11], conducted at EMPA only, showed that the partial flow system measured slightly higher values than the full flow system.
Similarly, according to [10], with the use of diesel particulate trap (CRT), partial flow dilution systems measured slightly higher PM. This was further confirmed by the findings of [11]. As shown in Figure 10 the partial flow system tended to measure slightly higher particulate emissions than the full flow system. The percentage difference was lower than 10 % for the engines without aftertreatment and increased up to 50 % for the engine with CRT.
Figure 10. PM emissions results for 3 different engines in worldwide harmonised test cycles WHTC (transient) and WHSC (steady). Note that engine 2 is equipped with CRT [11]
Volume 2
Low Levels of Particulates
Page 38
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
In [10], PM measurement accuracy was good down to PM levels of 0.015 g/kWh, when PM is mainly carbonaceous. Accuracy was significantly deteriorated, if the main portion was SOF and/or sulphate; this problem especially occurred with aftertreatment (CRT) systems.
In [11] the absolute difference between the systems was below 0,007 g/kWh for all engines tested, that is it remained the same with or without an aftertreatment system. Compared to the results of [10], the repeatability of PM measurement with CRT was much better. For all transient test cycles, the standard deviation was at or below 20% of the average value of three tests. In absolute values, the standard deviation was between 0,001 g/kWh and 0,003 g/kWh. Fuel free of sulphur (2 ppm) was considered as the major reason for this improved repeatability.
According to [14], PM measurement is possible with 10% accuracy down to PM levels 0,01g/kWh, since resolution is 0.001 g/kWh. However, measurement accuracy deteriorates, if the predominant portion of PM is sulfate and SOF. The problem is related to PM definition (exhaust dilution) rather than to the gravimetric method (weighing) and will also occur with any other PM measurement method. The sulfate problem will especially occur when using exhaust aftertreatment systems with oxidation catalysts. The PM measurement problem can only be avoided either if sulfur free fuel and oil are used, or sulfate is not considered part of PM. Table 5 and Table 6 show the accuracy of PM measurement based on regulatory requirements for steady and transient tests respectively. The relative and absolute variability of PM measurement are shown in Figure 11 and Figure 12, respectively.
In a recent publication Silvis et al. [16] report on the examination of possible error sources (deposition losses, sampling losses, errors due to gas composition, soluble fraction errors, time delay errors) and measures taken to minimise them. A partial flow dilutor incorporating these measures was tested in an EPA/ARB/EMA study [15]. As far as steady state tests are concerned, results show good agreement for low load modes (Figure 13). In higher loads deviation was observed, which was attributed to differences in the PM soluble fraction. The later was confirmed by the significant improvement obtained by appropriate matching the dilution ratio and mix temperature. The transient test results (Figure 14) show very good overall agreement (5% or below) with full flow system having the same or slightly better variability.
In addition to PM, these studies concluded that the raw gas measurement showed a good agreement to the diluted measurement. For nitrogen oxides and carbon dioxide, the differences between raw and diluted measurement were below 3 % for all test cycles and engines. For carbon monoxide, the relative differences were between 5 and 20 % for the engines without trap and higher for the low emitting engine. In absolute numbers, the differences were lower than 0.4 g/kWh, which is considered acceptable with respect to the CO emission standard. It was therefore suggested that, due to the higher gas concentrations in the raw gas, the ISO measurement procedure is advantageous for very low emitting engines, e.g. with aftertreatment systems. A clear improvement is the raw measurement for hydrocarbons: the repeatability of the measurement went down to half of the value with the diluted measurement. Volume 2
Low Levels of Particulates
Page 39
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Table 5. PM Accuracy Based on Regulatory Requirements (ESC tests) [14]
Error ± 3.9 % ± 4.0 % ± 2.0 % ± 2.0 % ± 10.0 % ± 12.0 % ± 5.9 % ± 6.3 %
Measured Value Mf (mg) GEDFW (kg/h) MSAM (kg) Power (kW) PM (g/h) PM (g/kWh) PM (g/h) PM (g/kWh)
Remarks Based on 2.5 mg particulate weight Max. error allowed Max. allowed error is ± 3.6 % = max. error = average error (RMS method)
Table 6. PM Accuracy Based on Regulatory Requirements (ETC tests) [14]
Error ± 4.2 % ± 2.0 % ± 4.9 % ± 2.0 % ± 11.2 % ± 13.1 % ± 6.8 % ± 7.0 %
Measured Value Mf (mg) MTOTW (kg) MSAM (kg) Cycle work (kWh) PM (g/Test) PM (g/kWh) PM (g/Test) PM (g/kWh)
Remarks Based on separate weighing Max. error allowed Based on double dilution = max. error = average error (RMS method)
ISO correlation study Engine with CRT 18 ppm S fuel
35%
30%
Measurement tolerance of procedure allowed by Regulations Weighing error is only 0.8 %
ISO correlation study Engine w/o CRT 18 ppm S fuel
WHDC validation study Engine with CRT 2 ppm S fuel
25%
20%
15%
10%
5%
0% 1999/96
ISO US 2007 16183
A 100
B 100
B 50
B 25
ESC
ETC
ESC
ETC
ESC
ETC
WHSC
WHTC
Figure 11. Relative PM variability of heavy duty engines [14]
Volume 2
Low Levels of Particulates
Page 40
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
0,07
ISO correlation study Engine w/o trap 18 ppm S fuel
0,06
ISO correlation study Engine with CRT 18 ppm S fuel
Level STD
WHDC validation study Engine with CRT 2 ppm S fuel
0,05 Possible resolution of gravimetric method: 0.001 g/kWh Possible resolution of weighing procedure: 0.0001 g/kWh
0,04
0,03
0,02
0,01
0,00 A 100
B 100
B 50
ESC
ETC
ESC
ETC
ESC
ETC
WHSC
WHTC
% Diffe re n c e re la tive to C VS
Figure 12. Absolute PM variability of heavy duty engines [14]
Ste a d y Sta te - % Diffe re n c e to C VS
1 5%
1 0% 5%
0%
A
B
C
D
E
F
-5% -10%
Figure 13. Steady state results (A: Rated speed, 100% load, B: Rated speed, 50% load, C: Rated speed, 10% load, D: Maximum torque engine speed, 100% load, E: Maximum torque engine speed, 50% load, F: idle) [16]
Volume 2
Low Levels of Particulates
Page 41
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
0 ,35 0
Gm /hp -hr
0 ,30 0 0 ,25 0 0 ,20 0 0 ,15 0 0 ,10 0 0 ,05 0 0 ,00 0 1
2
3
C VS PM Em issio ns
4
5
6
7
PFD PM Em issio ns
Figure 14. ETC transient results [16]
4.3 FILTER HOLDER TEMPERATURE As already mentioned, the US 2007 regulations introduce a narrow temperature range (42 to 52oC) for the PM filter holder. The introduction of this temperature range is the most advanced requirement of the US regulations, because it attempts to accurately control the organic fraction of the measured PM. This is especially valid in view of the possibility that diesel particulate traps are widely used for compliance with future emission standards, where SOF will constitute the dominant part of PM. This is clearly illustrated by the measurements reported in [10, 11 and 12] with the CRT systems (see for example Figure 10, where the relative-to-mean scatter of PM test results of the low emitting engine 2 is much higher than the more conventional engines). Additional equipment will be required for PM measurement, in order to simultaneously keep the two temperature limits imposed - primary dilution for gaseous emissions to a temperature below 191°C, secondary dilution for PM in the temperature range 42 – 52°C - without modifying the primary dilution tunnel flow rate between full load and idle. In principle there are several possibilities to keep the filter holder at this temperature range while keeping the dilution ratio at relatively low levels to enhance the repeatability of the measurements:
either heat up the filter holder and the sampling line (i.e. put the whole system into an oven of controlled temperature)
Volume 2
Low Levels of Particulates
Page 42
EC-DG ENTR
Emission control technology for heavy-duty vehicles
or heat-up the dilution air to a suitable temperature.
or a combination of the two.
ETD/00/503430
ISO 16183 allows dilution tunnel and sample pipe heating to a temperature up to 52°C. This option is also included in the US regulation, as an alternative to heavy insulation of the corresponding pipes, without specifying the maximum permissible temperature. Still, it has to be mentioned here that in the US regulation it is stated that: «Measurement procedure intent is to maintain sample temperature by dilution and mixing with air rather than by transferring heat to the surfaces of the sampling system», and this may be considered as contradictory to the possibility of heating the sample in the connecting pipe. Discussions with parties involved in these developments [17, 18, 19] indicated that i.
the US EPA proposal for the narrow temperature range has been well received because it should
reduce measurement uncertainty, ii.
the way this will be realised is still unclear, while US EPA still awaits for comments and input
for the technical details of the regulation iii.
the first priority will be to look at the heating the sample holder rather than the dilution air.
Nevertheless, it has to be emphasised that from a measurement point of view one should be very careful in using filter holder and sample hose heating to keep the sampling temperature at the desired range, in particular in cases where there is the need to heat the diluted exhaust from temperatures around 15oC (see also below the discussion on dilution air temperature) up to 42oC. Thermal gradients (associated with thermophoresis and other phenomena) may affect the physical character of particulates at the sampling point. It is unclear how important this process will affect the mass of particulates. However, other parameters may substantially be affected (such as number and distribution of particles), which must be carefully studied, in particular if other metrics for particulates are envisaged for the near future. It is certain that mass measurement systems as those discussed for legislative purposes will also be used for the measurement of other particle properties. Figure 15 shows the results of a calculation (assuming adiabatic conditions) which illustrates how the temperature range 42 to 52oC could be achieved with hot dilution air (42oC dilution air temperature) and double dilution. In this example, a first dilution with a DR = 8 at the rated mode results in a diluted exhaust temperature of about 110oC. A second hot dilution with a DR = 10 (or 15) leads to 48oC (or 46oC) final dilution temperature. At idle, the first dilution leads to temperatures around 42oC, irrespective of the DR used, while the second dilution keeps the temperature at 42oC. Figure 15 presents also a drawback of this approach: the power which is required for heating up the dilution air from 20 to 42oC is substantial (of the order of 50 to 60 kW for primary dilution ratios of the order of 8 to 10). This is both technically difficult to realise and very expensive. Irrespective if hot dilution will be used in full flow systems to achieve the temperature range of 42 to 52oC, the above analysis shows that hot dilution may be successfully used in partial flow dilution systems.
Volume 2
Low Levels of Particulates
Page 43
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Figure 15. Calculated temperature at filter holder assuming adiabatic mixing with hot dilution (42oC dilution air temperature). Assumed values for the calculations: Texhaust at rated = 500oC, Texhaust at idle = 100oC, secondary dilution DR 10 and 15.
4.4 DILUTION AIR TEMPERATURE AND HUMIDITY Neither the draft ISO 16183(*), nor the US requirements define any upper limit for the dilution air temperature. Instead they introduce a lower permissible dilution air temperature of 15oC. It is correct to leave the upper dilution air temperature to be defined by the system manufacturer. As discussed in the previous section, this may give an additional freedom degree for the achievement of the lower temperature limit at the sample holder. In addition a high dilution temperature will ensure lower water vapour partial pressure, which in turn more reliably prevents condensation on the sample filter A lower than 20oC dilution air temperature limit, on the other hand, is needed, to offer the possibility of lower dilution factors (hence increased mass of PM collected on the filter), preventing in parallel water condensation on the filter. The relative humidity of the dilution air may also affect the measurement result, since it adds water in the exhaust. One would expect therefore a more precise definition of this parameter (as it is the case for filter conditioning, for example). In the EU Directive and the draft ISO standard the dehumidification is simply permitted, while the US regulation does not mention anything.
(*)
ISO 16183 indirectly specifies a maximum dilution air temperature of 52°C only in section PSP.
Volume 2
Low Levels of Particulates
Page 44
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Another property of the dilution air not defined in either the Directive, or the US regulations nor the draft ISO standard is the particulate content of the dilution air. Since particulates are determined on a mass basis only, the dilution air particulate load may become significant, especially in view of the very low particulate emission of the future engines. It is considered therefore essential to prescribe in the regulations either the maximum permitted particulate load of the dilution air (i.e. compulsory use of filtered air for dilution) or a way to compensate it, as is the case in the US regulations and the EU Directive (note that this correction is compulsory in US regulations and only optional in the Directive). Still both maximum background PM level and correction for it may be necessary, in cases of very low emitting engines. In summary any future requirement relating to the dilution air should include limits for: (a) the minimum temperature, (b) the relative humidity and (c) the permissible particulate content and/or compulsory background correction.
4.5 PARTICULATE SAMPLING PROBE A major difference between the US regulation and the EU Directive is the diluted exhaust sample preconditioning, before it passes through the sample filter. A pre-classifier is required by the US regulation installed before the open tube facing upstream on the exhaust pipe centerline - to remove the coarse, mechanically generated particles (e.g., rust from the engine exhaust system or carbon sheared from the sampling system walls) from the sample flow stream while allowing combustion-generated particles to pass through to the filter. The preclassifier may be either an inertial impactor or a cyclonic separator. The preclassifier manufacturer’s 50% cutpoint particle diameter has to be between 2.5 µm and 10 µm at the volumetric flow rate selected for sampling of particulate matter emissions. On the other hand and in the same direction the draft ISO standard foresees four different possibilities: a)
open tube facing upstream on the exhaust pipe centerline
b) open tube facing downstream on the exhaust pipe centerline c)
multiple hole probe
d) “hatted” probe facing upstream on the exhaust pipe centerline (known from the passenger car EU Directive). According to ISO 16183, a pre-classifier is recommended only in conjunction with a sampling probe of the "open tube facing upstream" design, while the other three alternatives by their design are believed to prevent large particles from entering into the probe. It is recommended that the future European regulations address specifically the sample probe issue and the diluted exhaust sample pre-conditioning. However, it is noted that in the framework of ISO discussions, the European and Japanese ISO members are of the opinion that the design required in the USA (open probe) is not a good technical solution, whereas the US delegation has an opposite view [17]. Hence the four alternatives included in the ISO proposal. Therefore it is also suggested that it is necessary to experimentally check these alternatives against each other, prior to any final decision. Volume 2
Low Levels of Particulates
Page 45
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
4.6 SAMPLE FILTER MATERIAL A major difference between Directive 1999/96/EC and the other two procedures concerns the PM sample $ 12 '&( ) *+,
99% in the others. Obviously, the latter presents higher collection efficiency in smaller particles, thus increasing the overall representativity of the test result.
4.7 NUMBER OF SAMPLE FILTERS The backup filter, required in Directive 1999/96/EC, is the same type as the primary one, and as such does not expand the collection range to smaller particle sizes, and therefore it was initially expected to collect only the losses of the primary one. These losses however must be assumed to be very low, especially in view of the low backpressure increase permitted (see relevant section). In addition, the backup filter may only play the role of condensation surface, which in itself introduces additional inaccuracies. Having in mind the very low load expected on the secondary filter, the expected accuracy with which it is determined (weight difference between loaded and unloaded filters), is limited, especially in view of possible filter material losses in the holder and humidity effects. It is believed therefore that the use of the backup filter has an over all negative effect on the accuracy of the measurement. Consequently, it is believed that the elimination of the backup filter in both the draft ISO standard and the US regulation, combined with the higher filtration efficiency (see above), will increase accuracy.
4.8 SAMPLE FILTER LOAD A difference between the EC Directive and ISO is found in terms of minimum sample filter loading. In 1999/96/EC, a minimum loading is recommended, but there is no specification concerning the minimum actually required. The minimum recommended load is in the range of 0,56 – 0,73% of the weight of the unloaded filter, the exact value depending on filter size. In order to increase the particulate mass collected on the sample filter, a sufficient number of test repetitions with the same sample filter is permitted, at the manufacturer’s request. In the ISO proposal, the minimum required load is prescribed instead, and the requirement is in the order of 0,13% of the clean filter weight, i.e. about 1/5 of that recommended in 1999/96/EC. Additionally, in ISO there is the provision for more than tests on the same sample filter, in order to collect the minimum required mass, provided that the parties involved (i.e. the laboratory and the engine manufacturer) agree. Also in ISO, filter loading lower than the minimum required is allowed, provided that a higher nvolved agree to that.
!
Volume 2
Low Levels of Particulates
Page 46
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
The US regulation simply recommends an as high as possible filter loading, without neither giving any limit values nor mentioning anything for test repetition(s) on the same filter. In view of the face velocity limit established in all cases (80 cm/s – EC Directive, 100 cm/s – ISO and US regulation), combined with the low emission of the engine and the dilution ratio, the US regulation simple suggestion for maximum possible filter loading is the most adequate, since it may be impossible to achieve the minimum either required or recommended loads with one test run (this problem is highlighted in [11]). On the other hand, allowing for test repetition on the same filter (as many times as it is required in both 1999/96/EC and ISO proposal) on the engine manufacturer request (EC directive) or agreement (ISO) in principle can not be accepted. The decision on whether the test has to be repeated on the same filter should be made only by the testing laboratory, which has the responsibility of the test result accuracy. However, based on recent experimental findings [11], the accuracy improvement by test repetitions on the same filter is questionable: the more repetitions were made, the lower the specific emission got, as shown in Table 7. Therefore, and until this issue is clarified, the US regulatory approach is suggested.
Table 7. Filter loading and emissions depending on the number of runs on the same filter pair [11]
!" "#" $$ % "# ## % "&#
4.9 STAIN DIAMETER An advantage of Directive 1999/96/EC and the US regulation over the draft ISO standard is the definition of the stain diameter. The lack of such a definition in ISO permits the testing laboratory to measure with filters effectively much smaller than the minimum prescribed (47mm), while using any acceptable size, by simply modifying the filter holder, reducing thus the stain diameter. Comparing Directive 1999/96/EC and US regulation it has to be mentioned that the Directive’s approach is advantageous as the stain diameter is prescribed as a function of the filter size, while in the US regulation only one value is given – corresponding to the minimum permissible filter size.
4.10 SAMPLE FILTER CONDITIONING Filter conditioning before and after the test at a specific temperature (22±3°C) and relative humidity ranges (45% ± 8% or the equivalent 9.5 ± 3°C dew point) is in all three procedures. However, the US regulation specifies a tighter humidity control, i.e. a dew point of 9.5 ± 1°C. Evidently this is done to minimise the hu-
Volume 2
Low Levels of Particulates
Page 47
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
midity effect on the accuracy of the weighing, due to the possibility that the particulate layer of the loaded filter absorbs ambient moisture, especially in view of the very low mass collected. In addition to the temperature and humidity, the US regulations specify that microbalance and filter stabilisation environments shall be free of ambient contaminants (such as dust or other aerosols) that could settle on the particulate filters. Therefore, it is recommended that these environments be built to conform to the Class 1000 specification (or cleaner) as determined by Federal Standard 209D or 209E for clean room classification. An alternative recommendation would be to equilibrate and/or weigh the filters within a separate, smaller, particle-free, temperature and humidity-controlled chamber (i.e., ‘glove box’ or small environmental chambers such as the one described in [20]). Weighing facilities, consisting of a small room in which temperature, humidity and dust are controlled, have environmental controls that typically cost around 50,000 from several hundreds to several thousand 20]. At the other end, ‘glove-boxes’ may cost around 10,000 ctronics. Overall, the US regulation provisions of tighter humidity and dust control of the conditioning and weighing room is a clear improvement, even though it may increase both installation and operation costs.
4.11 SAMPLE FILTER WEIGHING d-
* !'''3'4312 $5+ !4!6) ! $5+
% ! % 75 % !
As already mentioned, the sample filter load accuracy of determination is not a question of only the balance resolution, since it is calculated as the weight difference between loaded and clean filter. The effect therefore of minor filter material losses in the filter holder, as well as of humidity absorbed on filter material and on the accumulated particles or even spill of some particles when removing the filter from the holder can become very significant, and the increased resolution of the balance used can not compensate for them. The only way to overcome such effects is to increase the filter load. Nevertheless, the use of a balance of the highest possible accuracy is definitely in the correct direction. Concerning the repeatability of reference filter weighing, used as an indicator of the conditioning room envi % $5+ 75 ! !'''3'4312 &( & # 8 !6 !!
low overall sample filter loads expected. The adoption of a glove-type box can offer significant cost reduction for maintaining the environmental quality during conditioning and weighing.
Volume 2
Low Levels of Particulates
Page 48
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
4.12 WEIGHING CONDITIONS All the procedures require the same ambient conditions for the weighing room (temperature 22±3°C, relative humidity 45%±8%). The main difference is that the draft ISO standard and the US regulation recommend the use of a neutraliser (imperative in the US case, only for unstable balance readings in the case of the draft ISO standard). Static neutralisers, such as Polonium-210 sources, shall be used to neutralise charge on a filter prior to each weighing. It is considered that the overall weighing procedure benefits from using the neutraliser. In the US regulation it is also permitted to have multiple weighing runs and then use the average result.
Volume 2
Low Levels of Particulates
Page 49
EC-DG ENTR
Emission control technology for heavy-duty vehicles
5 ALTERNATIVE MEASUREMENT METHODS PARTICULATE MATTER MASS MEASUREMENTS
ETD/00/503430
FOR
DIESEL
Particle characterisation requires measurements that are accurate, repeatable, and realistic representations of diesel engine operation conditions. Since the newer engines will be cleaner, it will be increasingly difficult to quantify PM emissions using the conventional gravimetric method. To overcome this problem several alternative methods have been developed. The most recent is that of Horiba that uses gas analysers to measure potentially as much as several micrograms of PM. In addition, with this method it is possible to simultaneously analyse the volatile organic fraction (VOF), soot, and sulphates. All techniques under consideration for legislative purposes provide only an integral of PM emission over a transient cycle. However, instantaneous data become increasingly important for various purposes including engine research and development, emission inventorying, atmospheric modelling, health effect studies etc., i.e. activities that have an influence on legislation (e.g. the Auto- Oil programmes and ‘Clean Air for Europe CAFE’). These data are typically collected within emission factor programmes, which often adopt the legislative test procedures. In this context, numerous attempts have been made to determine PM emissions in near real-time. Moreover, as [2] points out, the mass emission determination in a raw exhaust sample and the measurement of the exhaust gas mass flow rate is a state of the art procedure for light duty vehicle development on chassis dynamometers. There it is called modal analysis. However, it is usually done in conjunction with the mass emission evaluation on a full flow CVS with bag analysis where quality of the modal results can easily be verified by comparison to the CVS bag results. For the above reasons, and with a view to the future (Euro 5 and beyond), it was decided to look in detail in the alternative methods proposed for the measurement of particulate mass. In the sections of this chapter, the operation principle of the most promising methods is briefly reviewed and discussed. Firstly, the most recent alternative method developed by Horiba is presented followed by the methods aiming at determination of PM emission in real-time.
Volume 2
Low Levels of Particulates
Page 50
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
5.1 HORIBA NEW MEASUREMENT METHOD The new measurement technique developed by Horiba [21] uses vaporisation, oxidation and deoxidation as illustrated in Figure 16.
Figure 16. Horiba MEXA-1370PM measurement system configuration (adapted from [21])
Sample gas from the dilution tunnel is drawn through a quartz filter, according to conventional procedures. The particulate sample is collected on the quartz filter, the characteristics of which remain unchanged under high temperature. The filter is then placed into a furnace heated to 980°C in a N2 gas flow. SOF and sulphate in the particles are vaporised. After vaporisation the sulphates are reduced to SO2 and measured by an SO2 NDIR detector, the vaporised SOF is converted to CO2 and measured by an NDIR detector. Subsequently O2 is introduced into the furnace at constant flow and the remaining soot is oxidised to CO2 and measured by the CO2 detector.
Figure 17. Gas analysers signals of MEXA-1370PM [21] Figure 17 shows a model concentration profile of CO2 and SO2, which are generated in the process described above. The output of the CO2 analyser is integrated while introducing N2 gas to determine the mass of SOF, Volume 2
Low Levels of Particulates
Page 51
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
and is integrated while introducing O2 gas to determine the mass of soot. The SO2 analyser output is also integrated to determine the mass of the sulphate. So far evaluation of the performance of the system has been conducted and published [21] by the instrument manufacturer only. Based on these experimental data, the following conclusions are drawn:
The method is able to analyse SOF, soot, and sulphate mass contained in PM within 4 minutes.
The results are well correlated with those of gravimetric method as Figure 18 shows. It is worth mentioning that Figure 18 shows that the method has been successfully used for low PM mass.
The detection limit of this technique is theoretically down to few micrograms. According to the manu - 5+"- - - 6
However, since the major source of PM measurement variability is the dilution, sampling, and filter conditioning procedure and not the weighing process itself, the instrument seems to be rather a promising development tool for PM analysis and not necessarily an alternative to the existing gravimetric technique. Furthermore, possible sample artefacts due to different filter used should be evaluated (quartz filter usually show a higher collection of SOF's (sampling artifact) than teflon coated or pure teflon filters). In any case, a thorough evaluation of its capabilities by independent laboratories is needed. US EPA is currently testing the instrument.
Figure 18. Correlation of Horiba MEXA-1370PM with filter gravimetric method (using PM sample) [21]
Volume 2
Low Levels of Particulates
Page 52
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
5.2 REAL-TIME MEASUREMENT TECHNIQUES As already mentioned numerous attempts have been made to develop methods for real-time PM determination, either directly or indirectly. These include optical methods such as light-extinction and light absorption (photoacoustic spectrometry) and light scattering, pressure differential measurements, flame ionisation detection and inertial mass measurements methods (tapered element oscillating microbalance)
5.2.1 Optical Methods In optical methods the interaction of aerosol particles with the incident light serves as a basis for the real time measurement of particle concentration [22]. The operation principles of these methods are based on the optical phenomena that are a direct result of the scattering and absorption of light by the aerosol particles. Important features of optical methods include high sensitivity, nearly instantaneous response and the avoidance of physical contact with the particles.
5.2.1.1 Light Extinction
Aerosol particles that are illuminated by a beam of light scatter and absorb some of that light and thereby diminished the intensity of the beam [23]. This process is called extinction and deals only with the attenuation of light along an axis. Light-extinction methods using commercially and non-commercially available instruments (smoke meters and opacity meters) have been used extensively to monitor diesel PM emissions and there is thus considerable information in the literature [24, 25, 26, 27, 28, 29, 30, 31]. Their operation is based on the attenuation of a light beam by the smoke according to Beer’s law:
θ = −
= − −
(1)
where is the opacity, I and I0 are the transmitted and incident light beam intensities, L is the path length through a smoke of particle mass concentration M and AE is the specific optical extinction. The latter is the effective optical cross-section for attenuation of light per unit particulate mass and has units of m2g-1, whereas the product AEM is the total attenuation or extinction coefficient and has units of reciprocal length. If AE is constant, which is true if the emitted particles are constant in size, shape, density and chemical composition, then the opacity is uniquely related to the mass concentration. This technique is illustrated in Figure 19. Pinhole appertur
I
Io Light Source
Volume 2
Lens
Lens
Pinhole appertur
Aerosol
L
Low Levels of Particulates
Detector Lens
Page 53
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Figure 19. Schematic diagram of an extinction measuring apparatus Despite its excellent time response, simplicity and cost-effectiveness this technique is inferior to measuring diluted exhaust. Since the dilution process significantly influences the quality and composition of the collected particulates, smoke meter measurements generally do not provide a good correlation with the total particulate mass (problems mainly related with the condensed fraction). The majority of correlations between PM mass concentration and smoke measurements published in literature were derived from steady-state engine tests. Concerning transient tests, Japar and Szkarlat [28] have suggested a method which can provide real-time PM emissions based on a correlation between the total particulate mass and a response of the opacity meter located in the diluted exhaust. Krempl et al [32] developed a method for separate but practically simultaneous measurement of both hydrocarbon and PM mass concentration. This method was based on optical extinction measurements at distinct frequency bands in the infrared (IR)-region. The authors found considerable differences between steady state and transient tests both in PM and hydrocarbon emissions. They brought their measurements to total PM emissions, which compared well to those determined gravimetrically over the whole measurement range.
5.2.1.2 Light Absorption and Scattering
Light passing through an optical system can be attenuated by absorption and by scattering. A number of techniques have been devised to measure light absorption by aerosol. Many use the capture of aerosol on filters, followed by an optical measurement to determine aerosol light absorption. The aethalometer [33] is a real time version of such an instrument. It is mostly utilised for ambient monitoring and recently has been tested in diesel exhaust measurement [31]. This instrument continuously measures the amount of light transmitted through a quartz filter, while particles are being deposited on the filter. The rate of decrease of transmissivity, divided by the sample flow rate, is directly proportional to the light absorption coefficient of the particles. Newer instruments have been calibrated in terms of the difference of light extinction and scattering in a long-path extinction cell, for laboratory test aerosols. Since it is an ambient measurement instrument designed for low aerosol concentrations, a secondary dilution is necessary. Thereby, uncertainties in the value of secondary dilution ratio may introduce additional errors into its readings. A light absorption method that can measure aerosol without the use of filters, and can observe the aerosol closer to its natural state is the photoacoustic technique (Figure 20). It has been applied in exhaust monitoring for more than 2 decades [28, 29, 34, 31]. In this method electromagnetic energy absorbed by the aerosol is converted to heat. Because the aerosol is small, and has sufficiently high thermal conductivity, the absorbed heat will flow rapidly to the surrounding air. Heated air responds by expanding its volume and/or pressure. By placing the aerosol-laden air into an acoustic resonator, and modulating the electromagnetic power at its resonance frequency, the varying pressure disturbance (acoustic signal) can be amplified by the build-up of a standing acoustic wave in the resonator. Thus by measuring the sound pressure that is associated with aerosol light absorption, a measure of elemental carbon concentration can be obtained. Volume 2
Low Levels of Particulates
Page 54
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
The main conclusion derived from photo acoustic device application in diesel PM emissions measurement is that its sensitivity depends strongly on the flow rate of the sample gases, the composition of the diluting gases and the modulation frequency of the light. Another problem resulting from operation characteristics concerns the sensitivity of the technique to the percentage of volatile material present in the exhaust.
Figure 20. Schematic view of the prototype photo acoustic spectrometer
5.2.2 Pressure Differential Technique Pressure-differential measurement is simple and relatively inexpensive once a dilution system is available. It is based on the measurement of pressure drop increase with particle loading across the filter. However, its purpose is not to replace the dilution–filtration hardware, but to add the ability to obtain real-time information. The mass–pressure-drop correlation may be dependent on particle size and composition. The change in pressure drop per unit time and concentration has been shown to be a strong function of relative humidity, decreasing with increasing relative humidity. These results suggest that particulate concentration measurements that use the pressure drop method may be subject to additional uncertainties if used in an environment where the ambient relative humidity cannot be controlled accurately [35, 36].
5.2.3 Fast-Response Flame Ionisation Detection Flame ionisation detectors (FIDs) are widely used to measure gaseous HCs and to estimate the amount of condensed/adsorbed HCs on the particulate because of their fast response and reliability [37]. Moreover, these instruments have recently been used as a technique for the real-time analysis of PM [38, 39, 21]. The basic physical phenomenon that is exploited in a FID is the ionisation that occurs when hydrogen atoms of high kinetic energy collide with another atomic or molecular species, which are then ionised [38]. The collision produces positive ions and electrons, which move towards the electrodes by the application of an electrical field, thus producing the ionisation current. The detector is based on the fact that an H2-O2 flame produces relatively few ions (about 107 cm-3) but it does contain highly energetic atoms. Therefore, when
Volume 2
Low Levels of Particulates
Page 55
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
trace amounts of HC compounds are added to the flame, the number of ions increase to the order of 1011ions/cm-3 and hence a measurable ionisation current is produced. This current signal is proportional to the number of carbon atoms present in the flame per unit time. Fast response FID is differentiated from the conventional one in that the sample gas is introduced directly into the flame chamber to minimise the gas transportation time, which would lead to significant signal delay due to the gas diffusion and mixing processes in the flow path. Figure 21 and Figure 22 illustrate a schematic view of a FID and a typical signal, respectively.
Figure 21. A schematic view of FID [38]
Figure 22. A typical FID signal Arcoumanis and Megaritis [37] used a fast response FID together with a smokemeter targeting to the development of an alternative method of quantifying real-time PM concentration during steady and transient engine operation. The method was based on instantaneous measurement of smoke and unburned HCs in the Volume 2
Low Levels of Particulates
Page 56
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
raw exhaust and their steady state correlation with the PM mass concentration in the diluted exhaust. Steady state tests on a number of engines (both naturally aspirated and turbocharged) exhibited the form
P = 1.038C + 0.523HC for naturally aspirated engines and P = 1.044C + 0.784 HC for turbocharged engines where P: particulate mass in diluted exhaust, C: carbon mass concentration in the raw exhaust (conversion based on Bosch smoke units) and HC: hydrocarbon concentration in the raw exhaust (FID volumetric concentration converted into mass assuming 13,85 molecular weight). An estimate of technique accuracy, obtained by integrating the transient particulates signal over the testing period and comparing it with the PM mass collected on the filter during the same period, showed, in all tested cases, a difference within 20% (higher than the calculated mass concentration). For more reliable results, the authors suggested the use of two FIDs measuring rather than empirically determining the fraction of the HCs, which is adsorbed or condensed on the particulates. One FID would measure the total emitted HC in the raw exhaust sampled at 190°C and other the total emitted HC at or below 52°C by taking samples from the diluted exhaust. From the difference of the FID signals the particulate phase organic emissions could be determined. It has been shown [40] that the difference between the two FID measurements is in excellent agreement with the particulate SOF. However, when a soot particulate is brought into the FID flame, a sharp ion pulse is generated due to its attached HCs being chemi-ionised. The amplitude of the signal can be much higher than that of the gaseous HCs present in the exhaust. Sun and Chan [38] have postulated that if an adequate number of particulates were introduced into the FID flame, it would be possible to measure its number density by counting the number of ion pulses for a specific time duration, provided that the sampling flow rate is kept constant. Being an indication of the particulate emission, the particulate number density could then be measured in realtime and subsequently used for transient PM emission analysis. In addition, the HC fraction of individual particulates and the gaseous HC concentration of the exhaust could also be determined simultaneously.
Figure 23Experimental set-up of Sun and Chan [38]
Based on the above Sun and Chan [38] modified the transfer tube of a fast response FID to allow ion pulses to be generated and conducted measurements over steady state and transient operation. The standard filtraVolume 2
Low Levels of Particulates
Page 57
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
tion method was used as a means of comparison. The experimental set-up is shown in Figure 23. The particulate number density determined by the modified FID was compared with the mass concentration measured by the filtration method under various steady engine-operating conditions. The particulate number density rose as the PM mass concentration increased. However, the authors pointed out that the correlation between these two parameters would be possible only if the size distribution and mass density of PM were determined. In transient operation they observed a clear increase in particulate number density attributed to excursion of the engine into off-design operating conditions. As regards the HC fraction of the particulates, they found to be in close association with the gaseous HC concentration under steady engine conditions but they observed a clear discrepancy between the two parameters also attributed to off-design operating conditions. Kawai et al. [39] examined the use of a FID spike response as an indicator of PM concentration. They compared the FID response to HC-free carbon particulate with those of a standard PM generator and of an actual diesel engine and pointed out that, although difficult to determine the correlation between FID response and PM mass for the first example, the second and the third case showed a linear correlation indicating a strong potential for the use of this technique in diesel engine development and evaluation. For the organic volatile material measurement, following the suggestion of Arcoumanis and Megaritis,[37] they used a double FID system where the sample gas containing PM was split into two lines. One line was heated to 191°C and the sample passed directly into the FID device; on the other line the sample was filtered according to the standard regulation method in order to remove PM and the residual gaseous HC compounds were passed into the second FID. The difference between the two signals was due to HC compounds in the vapour phase at temperatures between 52°C and 191°C and soot that was removed by the filter at 52°C. Comparisons with volatile organic material measurements by the gravimetric method showed good correlation in general. Fukushima et al. [41] used the above mentioned differential technique for continuously measuring soot and volatile material separately. Over steady state tests exhaust PM was simultaneously measured by the FID system and by the dilution tunnel using the standard filtration method. The correlations were good for both soot and volatile material (better for soot). Plots of the direct (from the engine without dilution) measurement also yielded good correlations between the diluted filter and the raw exhaust measurements obtained by the real-time PM analyser. Over transient-operation tests the PM analyser gave the possibility to analyse soot and volatile material emissions indicating the variations in engine loads and speeds. Conclusively, the authors suggest that the system is a very powerful tool in diesel emissions research.
5.2.4 Tapered Element Oscillating Microbalance (TEOM) 9:;gravimetric mass) when masses were located off-centre on the filter with the largest one for masses nearest the filter edge. They pointed-out, however, that since individual filters behave differently, non-bias or negative biases are possible. Thirdly, they conducted tests for possible filter surface deformations due to downstream vacuum conditions and subsequent sources of biases and concluded that these have to be considered as important only if volatile masses are of interest. Having demonstrated that the TEOM with second-generation filters (Pallflex type instead of Balston type media) exhibited a highly linear reproducible response to mass on the filter, they proceeded to the direct comparison of TEOM total mass readings to conventional 47-mm filter mass determinations. Their main conclusion was that the TEOM vs. gravimetric difference in mass was due to the fact that water and organic volatile material are retained to a greater degree by conventional filters. This difference was attributed to the higher vacuum on a TEOM filter. They suggested a correspondence between this higher vacuum with a temperature increase in conventional filtration in order to achieve improved agreement. Volume 2
Low Levels of Particulates
Page 60
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Shore and Cuthbertson [44] tested various light and heavy-duty engines over different transient cycles in order to assess the suitability of the TEOM for PM mass measurements and to determine its accuracy by comparing the results with those obtained by conventional procedures. The vehicles tested yielded real-time data significantly different from one another. Although the authors were not in the position to pinpoint the engine parameters crucial in determining the shape of readouts, this finding is considered as important indicating TEOM as appropriate in particulate measurement programs during which specific areas of high particulate emissions can be identified. Differences observed during tests over different cycles were attributed to the driving schedule specifications. Extensive data were given regarding the correlation between TEOM results and those determined gravimetrically. The data showed a considerable degree of variation from one vehicle to another, one cycle to another and even from one test to another on the same vehicle over the same cycle with the results obtained by TEOM being lower. These discrepancies were attributed to differences in the filter temperature leading to lower retention of volatile material by the TEOM filter (operated at a constant temperature (50°C) due to temperature dependent frequency characteristics of the element) and to differences in density and structure of the two media, which could influence the applicability of the value used in TEOM data reduction. To check the value supplied by the manufacturer the authors used the single mass method and found this was correct within the experimental error. However, they also checked the calibration by injecting a known mass of dioctyl phthalate (DOP) on the surface of the collection filter. This suggested that the TEOM readouts were up to 10% lower (on average) when the value supplied by the manufacturer was used. However, the mass used in this experiment was unrepresentative (much higher) of those detected during particulate collection. The authors also observed that limiting injections to the centre of the filter influenced the measurements. These results suggest that if one elects to verify the value, depositions similar to those expected during sampling should be used. As regards the influence of sample flow rate they found that there was a near-direct proportion of filter retention to flow rate for flows within 0,8 to 3 l/min (the TEOM unit can be set for sample flow rates of 0,5 to 5 l/min) and concluded that a rate chosen within this range was not critical. Another possible influential factor mentioned by the authors was leakage of sample into the element housing. To avoid this, flow of purged air to keep the feedback optics clean is suggested. Saito et al. [45] conducted comparison measurements using TEOM, a high-sensitivity light extinction opacimeter and the standard filtration method. TEOM data were well correlated to the filter collection method under cruise and modal operating conditions. Comparisons with opacimeter measurements showed that the TEOM data were also correlated except for some distinctive engine conditions. In continuous measurements it was found that the opacimeter is better than the TEOM method in terms of transient response and sensitivity. In 1998 D.A. Okrent of Pupprecht & Patashnick Co. published a paper where the third generation of TEOM device was presented. Focusing on the reported factors thought to affect TEOM data, the author formulated a method that optimises the signal to the microbalance. He found that the filter/microbalance temperature has Volume 2
Low Levels of Particulates
Page 61
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
the greatest impact on the overall total mass of the sample collected, and tended to reduce the moisture and VOC adsorption rate. For this reason adequate warm-up times after a filter change is suggested to provide sufficient temperature stabilisation. Figure 26 shows the results obtained with the filter operating at 35°C in comparison with those obtained at 55°C. He also suggests the use of the TX40 Pallflex filters, which due to their higher efficiency, shorten the time needed to build up a particulate deposit on the filter. In contrast with the sample flow rate range suggested by Shore and Cuthbertson, Okrent [44,49] claims that a better correlation between TEOM and standard filtration method was found for higher flow rates in the range of 5 l/min. Although a high flow rate decreases the filter life, this finding, if confirmed by further investigation, is of great importance since high flow rates are necessary for the latest technology of clean diesel engines. Consequently, filter life should be sacrificed in the interest of obtaining better quality data.
Figure 26. TEOM readouts for operation temperatures 35 and 55°C
Chan and He [47] introduced a new technique for real-time measurement of PM mass concentration using a third generation TEOM and a fast-response FID. This combination was judged as necessary to compensate for HC component, since neither TEOM nor other instruments for mass measurement concentration have been able to fully capture the HCs signature/component in the diesel exhaust. Maintaining TEOM collection temperature at its standard value (50°C), they derived an empirical formula for the correlation of mass of particulate emissions such as
= +
(3)
where , and are all in mgm-3, while the coefficients 0,981 and 0,257 showed the weighted contributions of and in , respectively. According to this correlation, an average of 25,7% by mass of HC is recovered by FID measurements. Volume 2
Low Levels of Particulates
Page 62
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Their results showed that both TEOM and FID respond rapidly to variations in diesel exhaust content and, consequently, they have to be considered as usable for real-time PM mass concentration measurement. Jarrett et al. [48], recognising adsorption-desorption of moisture as an important mechanism influencing TEOM accuracy and sensitivity, developed a model to predict its effect in the TEOM filter. Their results showed a strong linear correlation between moisture mass on the filter and ambient moisture content in the air. Conducting measurements with new and used filters concluded that the later adsorbed more moisture. This finding suggests that filter moisture retention is a function of accumulated PM on the filter. They suggested subtraction of the model predicted change in moisture mass from the initial TEOM data in order to yield more accurate representation of legitimate instantaneous PM and to prepare more accurate emission inventories. Moosmüller et al. [31] compared five different methods for measuring PM mass: light scattering method (nephelometer, DT); two light adsorption method, one that measures light absorption of aerosol deposited on a filter (aethalometer) and one that measures light absorption in situ (photoacoustic instrument, PA); one light extinction method (smokemeter, SM); and the TEOM method. Two modern vehicles and an older one were tested according to the FTP driving cycle. The averaged PM mass measurements obtained from each method (except aethalometer) for the modern vehicles are shown in Table 8 together with those from the conventional method. Table 8. Reproducibility of PM mass measurements [31]
CFR Filter [mg/m3]
TEOM [mg/m3]
TEOM together with nephelometer gave the best correlation with the standard filtration method in an older engine as well as in modern engines. It should be mentioned, however, that the TEOM regression equation of the modern engines is comparable with that of the older engine, whereas nephelometer regression equation of the older engine had a slope more than two times smaller than those of the modern engines. They concluded that each of these two instruments has distinctive strengths: the TEOM excels in the area of constant calibration, independent of vehicle, relative to the filtration method, whereas nephelometer provides an excellent signal-to-noise ratio, freedom from interference due to other sample properties, good time resolution,
Volume 2
Low Levels of Particulates
Page 63
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
simplicity, and low price. Considering these strengths to be complimentary, they suggested the operation of the two instruments in parallel in order to calibrate nephelometer signal in near real time using TEOM.
5.2.4.1 LAT in-house Experience with TEOM
PM mass measurements using TEOM were conducted at LAT in the framework of DGTREN “Particulates” Project, using as test vehicle a Diesel Euro I VW Golf.
Figure 27. “Particulates” measurement set-up A schematic of the particulate measurement set-up is presented in Figure 27. A small portion of the exhaust gas enters the primary dilutor and is diluted with dehumidified (humidity # ? should be contrasted to the cost of the new Horiba MEXA-1370 PM which is currently in the range of 200 k. Nevertheless, one should note that
this cost could be reduced to half if the instrument is in series production (the method is still considered to be experimental)
the instrument offers the simultaneous measurement of SOF, sulphates and insoluble fraction of particulates, which is tedious and time consuming with the conventional techniques
the use of the instrument offers substantial cuts in operational costs, as apparently the process is extremely fast compared to the conventional gravimetric techniques.
Finally a survey of the costs of existing instruments for real time measurement of particulates indicated a large price range (up to 100 k" For example the current version of TEOM for Diesel emissions is offered at a cost of the order of 40 k
Volume 2
Low Levels of Particulates
Page 70
EC-DG ENTR
7
Emission control technology for heavy-duty vehicles
ETD/00/503430
NATURAL GAS FUELLED VEHICLES
Compressed natural gas (CNG) is an alternative fuel with the potential of meeting tighter heavy-duty engines’ emissions requirements. It is extracted from underground reservoirs, is a fossil fuel and is composed primarily of methane, along with other hydrocarbons including ethane, propane and butane, and inert gases such as carbon dioxide, nitrogen, and helium. Due to its composition it burns more completely than diesel fuel and produces fewer complicated hydrocarbons during the combustion process. Therefore, interest of using natural gas as an alternative fuel has increased in recent years, particularly in urban areas, because it offers the potential of reducing exhaust emissions. Although much effort has been devoted to the measurement and characterisation of particle emissions from diesel engines, there have been few investigations focused on the corresponding emissions from CNG fuelled spark ignition engines. Particle emissions, by weight, are substantially lower from a natural gas engine than from a diesel engine, as it has been shown in a number of studies conducted either in mobile or stationary emission testing facilities. The mobile laboratories utilise full sized dilution tunnels with critical flow venturies to dilute the raw exhaust before sampling. Typical dilution flows were 30 to 60 m3/min. The emissions instrumentation included particulate sampling units consisting of 70 mm filters bed with a secondary dilution tunnel with mass flow control. Filters are weighed in a controlled environmental chamber, which was housed, among the other instrumentation, in a trailer. The dilution tunnel and the related equipment were mounted either on the roof or on the ceiling of the trailer. In the most extensive study conducted in United States with a mobile laboratory [59], 150 vehicles (including trucks, buses and tractors) operated by 18 different transit authorities or fleet operators located in 11 states were selected to provide an overview of the vehicles in service. 10 different types of engines and 6 different types of fuels were tested. PM emissions characteristics were examined according to engine age, model, driving pattern, frontal area, transmission type, usage, fuel type, inertia, aftertreatment device, and environmental conditions (humidity, temperature, barometric pressure and altitude). The basic driving schedule used in these tests was the Central Business District (CBD) cycle. Fourteen identical segments are in this cycle. Each segment takes 40 seconds, which includes 10 seconds for acceleration, 18,5 seconds for constant speed (32,2 km/h) 4,5 seconds for deceleration, and 7 seconds for idling. Comparison of PM emissions of 8 CNG vehicles with 4 vehicles fuelled with low sulphur fuel showed an average of 0,027 grams/km for CNG versus 0,965 grams/km for diesel vehicles (97% reduction). In another study conducted the same year [60] included buses tested over the CBD cycle with various fuels according to Federal Regulations [61]. PM samples were collected by drawing exhaust through primary and secondary dilution tunnels, and passing the diluted exhaust through filters. PM emissions were on average 0,025 g/mile for CNG vehicles equipped with a catalyst and 0,05 for CNG vehicles without a catalyst, whereas for low sulphur diesel vehicle the corresponding emissions were 1,23 g/mile (97 and 96 % reduction, respectively).
Volume 2
Low Levels of Particulates
Page 71
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Measurements conducted according to the test procedure specified by the California Air Resource Board (CARB) [62] showed PM emissions for the CNG vehicle to be 0,01 g/kW-hr, whereas a typical diesel engine had a value of 0,27 g/kW-hr (96% reduction). Dilution and sampling were conducted according to the Federal Regulation using a mobile laboratory. Fritz and Egbuonu [63] measured various heavy-duty trucks that were converted to operate on CNG fuel and compared the results with those obtained when the trucks were equipped with gasoline and diesel engines. Each heavy-duty vehicle was equipped with its own exhaust sampling system configuration, chassis dynamometer tie-down adjustment and set up. The driving cycle used was the Urban Dynamometer Driving Schedule for heavy-duty Vehicles. The reduction in PM emissions after the conversion of diesel vehicles to CNG was in the range of 50%. Emissions testing of on-road buses in Boulder, Colorado [64], on the CBD driving cycle demonstrated a 97% reduction in particulate matter emission and a 58 per cent reduction in nitrogen oxide emissions when compared to diesel buses. In the most recent comparative study [65] PM emissions from CNG fuelled buses measured from 0,007 to 0,041 grams per mile (0.004 to 0.026 g/km), whereas these of diesel buses ranged from 0,15 to 0,32 grams per mile (0.09 to 0.2 g/km). Despite its PM emissions benefit, CNG has some drawbacks. Apart from the increased greenhouse and formaldehyde emissions [65, 63] preliminary research has shown that CNG engines may emit more ultrafine PM and researchers are now beginning to focus on the latter issue, with results favouring diesel engines [66, 67]. However, the evidence is contradictory. According to the US Department of Energy (DOE) [68] particle size distribution measurement and ultra fine particle counting are developing technologies and initial data is considered as mixed. Therefore, considerable attention is required before a final conclusion is drawn. The interest groups against the use of CNG completely ignore the fact that while particles come from all kind of combustion sources, it is their toxicity that should guide the control of emissions. Across the world, scientific studies have established that particulate matter from diesel exhaust is extremely toxic. It comprises particles coated with toxic polycyclic aromatic hydrocarbons (PAH), some of which are known to be the most potent carcinogens. Compared with diesel vehicles, CNG vehicles emit negligible amount of particles. Moreover, even the little particles that are emitted by CNG vehicles are not as toxic as particles emitted by diesel vehicles as CNG is composed of mainly methane. The DOE paper observed that CNG buses consistently emit dramatically less particulate matter than diesel buses. The trace amount of particulate matter associated with CNG is attributed to crankcase lubricating oil consumption (which also occurs in diesel engines). The DOE report said, “Some tests have shown that CNG actually produces much fewer ultrafine particles than diesel fuel. However, the study of particle size distribution measurement and ultra fine particle counting are developing technologies, and initial data is mixed. According to Weaver and Balam [69] even deterioration of the natural gas engine does not have significant effect on particulate emissions. Particulate matter emissions from natural gas engines are unlikely to increase substantially due to wear or inadequate maintenance - at least until the piston rings, valve seals, or turbocharger oil seals are so worn that oil control is lost. According to the authors, this is not surprising, since Volume 2
Low Levels of Particulates
Page 72
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
particulate matter emissions from natural gas engines are derived from lubricating oil rather than fuel combustion. On the other hand, some published data indicate [70, 71] that CNG PM has a higher percentage of PAH on a weight basis than diesel PM. So, if the emissions from the two engines were compared on an emissions weight basis, the CNG engine will have more carcinogens attached to PM than the diesel vehicle. It is obvious from all the above, that CNG vehicles emit significant lower mass of PM than diesel. Thus, specific attention is required during the exhaust dilution since the accurate measurement of PM mass is becoming a highly demanding task. Practically all problems mentioned in the previous chapters concerning the accurate measurement of low emitting diesel engines using the full flow dilution method are also valid for CNG engines as well. Thus, it is proposed that techniques and methods developed for low PM emitted diesel vehicles engines should also apply to CNG fuelled. Furthermore, attention should be given in the composition of the collected CNG PM mass in order to clarify the above mentioned contradictory findings published so far.
Volume 2
Low Levels of Particulates
Page 73
EC-DG ENTR
8
Emission control technology for heavy-duty vehicles
ETD/00/503430
SUMMARY AND CONCLUSIONS
The aim of this part of the study was to compile information describing and providing costs of improved emission sampling and measurement techniques, that can be used for the measurement of the very low levels of particulate mass emissions from engines meeting the Euro 4 and Euro 5 emission limits. The study looked in parallel at the two main elements of the emission testing procedures for low PM emitting engines: (a) the sampling systems and (b) the measurement systems. The dilution and sampling procedures in the Directive 1999/96/EC has adopted a number of possible configurations. For steady state cycles these include raw exhaust sampling for gaseous pollutants, partial flow dilution for PM and full flow dilution for both gaseous and PM emissions. For transient cycles they include full flow dilution for both gaseous and PM emissions, accepting in parallel the partial dilution with the approval of Technical Service. Practically all type approvals granted on the basis of this directive have been based on measurements with full flow CVS, as no detailed specifications exist for partial flow systems and raw exhaust measurements over transient cycles. Objective of the new ISO standard (ISO/FDIS 16183 - still under voting) was to close this gap, developing the necessary specifications for the measurement of gaseous emissions from raw exhaust gas and of particulate emissions using partial flow dilution systems under transient test conditions. Correlation studies were undertaken in the framework of this activity to unveil the relationship between the conventional CVS technique and partial flow dilution. In addition, these studies looked to a number of details of the conventional gravimetric measurement techniques, introduced by the ISO/FDIS 16183. Data from these correlation studies indicate that the partial flow systems are sufficiently developed and thus able to keep the necessary proportionality to exhaust mass flow rate as well as a constant dilution rate over the transient operation conditions. As regards PM emissions an agreement in the range of ±10% was established between full (according to 1999/96/EC) and partial flow dilution systems in the framework of WHDC GRPE activity. Recent improvements of sampling probe and mini dilution tunnel led to minimisation of this variability to lower than 5%. These developments clearly indicate that the equivalency criterion prescribed in the Directive 1999/96/EC can be met by partial dilution systems. An equally good agreement was found between raw and diluted measurement of gaseous components. In addition, these studies have clarified a number of important elements regarding PM filter handling and weighing, in view of the low emitting engines of the near future. In this context no trend of the PM level over filter face velocity and filter loading was found, when these parameters were varied within the ISO standard’s ranges. In contrast, increase of the filter loading through repetition of test runs on the same filter pair turned out to be questionable. For the parameters related to partial flow dilution systems, such as sample line temperature, length and diameter as well as tunnel heating, significant effects were observed only in a few cases. PM levels tended to be slightly lower with a longer sample line, while for sample line temperature, tunnel heating and sample line diameter the current ISO requirements seem to be acceptable. In addition dilution air temperature variation between 20 and 30oC was not found to affect the PM results significantly.
Volume 2
Low Levels of Particulates
Page 74
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
Statistically, the most significant parameter was the dilution ratio for both full flow and partial dilution systems. As the draft ISO/FDIS 16183 correlates well with the procedures established by Directive 1999/96, it can be considered as an alternative method for the measurement of PM over transient conditions and could be incorporated into a future Directive. Moreover, all ISO provisions regarding new sample probe designs as well as PM filter handling with the conventional weighing technique, represent a clear improvement towards the more accurate measurement of low PM emissions, with the exemption of repetition of test runs on the same filter, which should be avoided. On the other hand, the US EPA in January 2001 adopted new regulations for PM measurement for the 2007 emission standards. These regulations concentrate on the full flow CVS technique without adopting the partial flow systems. However they refine PM measurement first by introducing a narrow temperature range (42 to 52oC) at the PM sampling holder and secondly by adopting a particle preclassifier to remove coarse, mechanically generated, particles from the sample flow stream. They also introduce a number of further refinements regarding handling and weighing of the filters, as well as thorough preconditioning of the dilution air. Clearly the control of the filter holder temperature is the most advanced requirement, since it attempts to control the organic fraction of the measured PM, improving the overall repeatability of the tests. This is especially valid in view of the possibility that diesel particulate traps are widely used for compliance with future emission standards, where SOF will constitute the dominant part of PM. It is not yet clear how this requirement will be met. Heating of the filter holder and sampling line seems to be the preferred cost-effective approach. However, hot dilution (i.e. heating the dilution air for both primary and secondary dilution to temperatures close to 40oC) may also be an alternative possibility, but it is associated with high installation and operation costs in full flow dilution systems. It should be possible to adopt the US requirements in into the EC legislation, for both full and partial flow dilution systems, with the partial flow systems probably being more easily adaptable. However, there is the need for some targeted correlation studies to reveal the significance of the effects and the additional needs for existing systems. Regarding the PM measurement techniques, all investigations published so far concentrate mostly on the further development and accurate control of the conventional gravimetric approaches (filter + microbalance). However, one instrument manufacturer (Horiba) has developed a device that splits total PM into a nonsoluble fraction (carbon), a soluble fraction (mainly heavy hydrocarbons) and sulphates. First results reported by the manufacturer indicate a very good correlation with standard gravimetric techniques with detection
The
instrument still needs to be further validated by independent laboratories and regulatory bodies. It should be emphasised that all the techniques under consideration for legislative purposes provide only an integral of PM emission over a transient cycle. However, more and more instantaneous data are needed and used for different purposes including engine research and development, emission inventorying, atmospheric modelling, health effect studies etc., i.e. activities that have influence on legislation (e.g. Auto-Oil programmes and CAFÉ). These data are collected within emission factor programmes, which adopt to the largest possible extent the legislative test procedures. In this respect it was thought of interest to look at alternaVolume 2
Low Levels of Particulates
Page 75
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
tive instrumentation that offers the possibility for real-time PM measurement. In this area the study concludes that the Tapered Element Oscillating Microbalance (TEOM) may be one instrument that has a wide potential in this area. TEOM is the only instrument that measures mass inertia and not via an aerosol parameter correlated with mass. If we consider that the current approach for PM measurement is to minimise the loss of volatile material than to remove most of it from the measurement process, the operation of TEOM monitors at 30 or 40°C would be recommended in conjunction, if possible, with a nafion dryer according to the sample equilibration system.
With regard to the cost of the sampling and dilution systems the following can be concluded: The cost of partial flow systems ranges between 110 and 150 k
i.
!
dilution system, capable of measuring engines up to 500 kW and complying with Directive 1999/96/EC. The partial flow dilution systems provide an excellent low-cost possibility for new emission test laboratories. ii.
The additional requirements of the US 2007 regulations will increase both the installation and
operating cost of the full flow system. It is estimated that this cost increase will be in the range of 10 to 20% of an existing installation, assuming that the temperature control at the filter holder will be realised via heating the sample hose and the filter holder. iii.
The cost of the new Horiba PM instrument (MEXA-1370 PM) is in the range of 200 k
should be compared to 65-70 k ! 0.1 g readability, plus the cost of an up-to-date weighing room. Nevertheless, at this cost, MEXA-1370 PM offers additional information for the split of PM into three different fractions, which is considered an advantage. iv.
Current versions of TEOM for Diesel emissions have a cost of the order of 40 k
With regard to gas fuelled heavy-duty vehicles, a detailed literature survey showed that there are some data on PM emissions from CNG vehicles complying with earlier emission standards. These data, all measured using techniques originally developed for diesels, indicate that PM emissions from CNG engines are in the order of 10% of those from diesel engines of an equivalent technology, being, however, of a completely different chemical composition (mainly heavy hydrocarbons, with high amounts of PAHs). Taking into account that Euro 4 and 5 diesels will be associated with PM emission levels close to the CNG, it is proposed that the current developments in PM emission measurement should equally apply to gas fuelled vehicles as well. Nevertheless, more research is needed in this area in order to identify eventual problems and particularities.
9
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the thorough review of Mr. Jürgen Stein (DaimlerChrysler Powersystems). Special thanks also go to Dr. Claire Holman for the time she took to upgrade the language of this report.
Volume 2
Low Levels of Particulates
Page 76
EC-DG ENTR
10
Emission control technology for heavy-duty vehicles
ETD/00/503430
REFERENCES
[1] Directive 1999/96/EC of the European Parliament and of the Council of 13 December 1999
“On the approximation of the laws of the Member States relating to measures taken against the emission of gaseous and particulate pollutants for compression ignition engines for use in vehicles, and the emission of gaseous pollutants from positive ignition engines fuelled with natural gas or liquefied petroleum gas for use in vehicles and amending Council Directive 88/77/EC” [2] ISO/TC 22/SC 5/WG 2 N 195E
“ISO/FDIS 16183: Heavy-duty engines – Measurement of gaseous emissions from raw exhaust gas and of particulate emissions using partial flow dilution systems under transient test conditions”, Date 200202-07. [3] Federal Register: January 18, 2001, Volume 66, Number 12, Rules and Regulations “Control of Emissions from New and In-Use Highway Vehicles”
[4] D.B. Kittelson and J.H. Johnson “Variability in Particle Emission Measurements in the Heavy Duty Transient Test” SAE Paper 910738, Detroit USA 1991. [5] J.McLeod, D. Nagy, P. Schroeder, S. Thiel, M.A. Dearth, A.D. Colvin, T. Webb, K.R. Cardunner, D. Schuelze, R. Middleton, and A.M. Schlenker “A sampling system for the measurement of precatalyst emissions from vehicles operating under transient conditions” SAE Paper 930141, Detroit USA 1993 [6] W.M. Silvis “Mini-Dilution Sampling systems for Vehicle Exhaust Emissions Measurement” Autotest’96, Barcelona Spain 1996 [7] K. Engeljehringer and W. Schindler “Experiences with a Mini-Dilution System for engine Homologation and Development” SAE Paper 942418, Detroit USA 1994 [8] N. Hirachouchi , I. Fukano, and H. Nagano “Measurement of Unregulated Exhaust Emissions from Heavy Duty Diesel Engines with Mini-dilution Tunnel” SAE Paper 900643, Detroit USA 1990 [9] W.M. Silvis, R.Neal Harvey, A.F. Dageforde “A CFV Type Mini-dilution Sampling System for vehicle Exhaust emissions Measurement” SAE Paper 1999-01-0151, Detroit USA 1999
Volume 2
Low Levels of Particulates
Page 77
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
[10] H. Juergen Stein “Worldwide Harmonized Heavy Duty Emissions Certification Procedure. Exhaust Emissions Measurement ISO 2nd Interim Report” Informal Document, 41st GRPE session, January 2001 [11] T. Schweizer, C. Havenith, and H. Juergen Stein “Worldwide Harmonized Heavy Duty Emissions Certification Procedure. Validation Results” Informal Document, 43rd GRPE session, January 2002. [12] T. Scweitzer and J. Stein “A New Approach to particulate Measurement on Transient Test Cycles: Partial Flow Dilution as Alternative to CVS Full Flow Systems” SAE Paper 2000-01-1134, Detroit USA 2000 [13] H. Steven
“Development of a worldwide harmonized heavy-duty engine emissions test cycle-final report” Informal Document, 42nd GRPE session [14] J. Stein
“Potential of Gravimetric PM Measurement Method” Presentation to the Worldwide Harmonized Heavy Duty Certification Group, 43rd GRPE session, January 2002. [15] I. Khalek “Performance of Partial Flow Sampling Systems versus Full Flow CVS on PM Emissions under Steady state and Transient Operation” Report on results of EPA/ARB/EMA Study, December 2001 [16] M. Silvis, G. Marek, N. Kreft, and W. Schindler “Diesel Particulate Measurement with Partial Flow Sampling Systems: A New Probe and Tunnel Design with Full flow Tunnels” SAE Paper 2002-01-0054, Detroit USA 2002 [17] Personal communication with Mr. Jürgen Stein, DaimlerChrysler. [18] Personal communication with Mr. Miro Janda, Horiba Europe GmbH. [19] Personal communication with Mr. Thomas Schweizer, EMPA.
[20] R. Allen, M. Box, L-J S. Liu, and T.V. Larson “A Cost-Effective Weighing Chamber for Particulates Matter Filters” Journal of the Air &Waste Management Association, 51, 2001 [21] H. Fukushima, H. Uchicara, I. Asano, M. Adachi, S. Nakamura, M. Ikeda, and K. Ishidsa “An Alternative Technique for Low Particulate Measurement” SAE Paper 2001-01-0218, Detroit USA 2001.
Volume 2
Low Levels of Particulates
Page 78
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
[22] K. Willeke, and P.A. Baron “AEROSOL MEASUREMENTS. Principles Techniques and Applications” Van Nostrand Reinhold, N.Y., 1993 [23] W.C. Hinds “Aerosol Technology. Properties, Behaviour, and Measurement of Airborne Particles”, John Wiley & Sons, Inc, 1982 [24] A.W. Carey “Steady-State Correlation of Diesel Smoke Meters-an SAE Task Force Report” SAE Paper 690492, Detroit USA 1969 [25] F.J.Hills, T.O. Wagner, and D.K. Lawrence “CRC Correlation of Diesel Smokemeter Measurements” SAE Paper 690493, Detroit USA 1969 [26] C. T Vuk, M. A. Jones, and J. H. Johnson “The Measurement and Analysis of the Physical Characteristics of Diesel Particulate Emissions” SAE Paper 760131, Detroit USA 1976 [27]G. Greeves, and C. H. T. Wang “Origins of Diesel Particulate Mass Emissions” SAE Paper 810260, Detroit USA 1981 [28] S. Japar, and A. Szkarlat Real-Time Measurements of Diesel Vehicle Exhaust Particulate Using Photoacoustic Spectroscopy and Total Light Extinction” SAE Paper 811184, Detroit USA 1981 [29] D. M. Roessler,
“Photoacoustic Insights on Diesel Exhaust Particle” Applied Optics. 23, 1148-1155, 1984 [30] J.S. Alkidas, “Relationships Between Smoke Measurements and Particulate Measurements” SAE Paper 840413, Detroit USA 1984 [31] H. Moosmuller, W.P. Arnott, C.F. Rogers, J.I. Bowen, J.A. Gillies, W.R. Pierson, J.F. Collins, T.D. Durbin, and J.M. Norbeck “Time Resolved of Diesel Particulate Emissions. 1. Instruments for Particle Mass Measurements” Environmental Science & Technology, 35, 781-787, 2001 [32] P. Krempl, W. Schindler, and E. Schiefer “Dynamic Measurements of Particulate Mass Emission on Light-Duty Diesel Engines Under SteadyState and Transient Conditions” SAE Paper 850269, Detroit USA 1985
Volume 2
Low Levels of Particulates
Page 79
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
[33] A.D.A Hansen, H. Rosen, T. Novakov
“The Aethalometer-An Instrument for the Real-Time Measurement of Optical Absorption by Aerosol Particles” The Science of the Total Environment 36, 191-196, 1984 [34] T. Abe, T. Sato, and M. Hayashida, M. “Particulate Matter Emission Characteristics under Transient Pattern Drivings” SAE Paper 890468, Detroit USA 1989 [35] H. Nogouchi, N. Mori, and T. Inoue “A New Practical Technique for Real-Time Particulate Mass Monitoring” 3rd Symposium on Advances in Particle Sampling and Measurement, Florida 1981 [36] T. Inoue, H. Nogouchi, K. Aoki, N. Mori, and S. Ikeda “Particulate Emission Characteristics from IDI Diesel Engine under Transient Operation” SAE Paper 820024, Detroit USA 1982 [37] C. Arcoumanis, and A. Megaritis “Real-Time Measurement of Particulate Emissions in a Turbocharged DI Diesel Engine” SAE Paper 922390, Detroit USA 1992 [38] J.H. Sun, and S.H. Chan “A Time-Resolved Measurement Technique for Particulate Number Density in Diesel Exhaust Using a Fast-Response Flame Ionization Detector”, Measurement Science and Technology, 8, 279-286, 1997 [39]T. Kawai, Y. Iuchi, S. Nakamura, K. Ishida “Real-Time Analysis of Particulate Matter by Flame Ionization Detection” SAE Paper 980048, Detroit USA 1998 [40] M.K. Abbass, G.E. Andrews, P.T. Williams, and K.D. Bartle (1989) “Diesel Particulate Composition Changes Along an Air Cooled Exhaust Pipe and Dilution Tunnel” SAE Paper 890789, Detroit USA 1989 [41] H. Fukushima, L. Asano, S. Nakamura., K. Ishida, and D. Gregory “Signal Processing and Practical Performance of A Real-Time PM Analyzer Using Fast FIDs” SAE Paper 2000-01-1135, Detroit USA 2000 [42] R. Whitby, R. Gibbs, R. Johnson, B. Hill, S. Shimpi, and R. Jorgenson “Real-Time Diesel Particulate Measurement Using a Tapered Element Oscillating Microbalance” SAE Paper 820463, Detroit USA 1982 [43] R. Whitby, R. Johnson, and R. Gibbs “Second Generation TEOM® Filters-Diesel Particulate Mass Comparison between TEOM® and Conventional Filtration Techniques” SAE Paper 850403, Detroit USA 1985
Volume 2
Low Levels of Particulates
Page 80
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
[44] R.P. Shore, and R.D. Cutbertson “Application of a Tapered Element Oscillating Microbalance to Continuous Diesel Particulate Measurements” SAE Paper 85045, Detroit USA 1985 [45] K. Saito, and O. Shinozaki “The Measurement of Diesel Particulate Emissions with a tapered Element Oscillating Microbalance and an Opacimeter” SAE Paper 900644, Detroit USA 1990 [46] H. Burtscher, S. Kunzel, and C. Huglin “Characterization of Particles in Combustion Engine Exhaust” Journal of Aerosol Science, 29, 389-396, 1998 [47] S.H. Chan, and Y.S. He “Measurements of Particulate Mass Concentration Using a Tapered-Element Oscillating Microbalance and a Flame Ionization Detector” Measurement Science and Technology, 10, 323-332, 1999 [48] R.P. Jarrett, N.N. Clark, M. Gilbert, and R. Ramamurthy “Evaluation and Correction of Moisture Adsorption and Desorption from a Tapered Element Oscillating Microbalance” Powder Technology 119, 215-228, 2001 [49] D.A. Okrent, “Optimization of a Third Generation TEOM® Monitor for Measuring Diesel Particulate in Real-Time”, SAE Paper 980409, Detroit USA 1998 [50] http://www.rp.co.com
TEOM Series 1105 Brochure, [51] H. Patashnick, E.G. Rupprecht. “Continuous PM10 measurements using the tapered element oscillating microbalance” Journal of Air and Waste Management Association, 41, 1079-1083, 1991 [52] J.C. Chow “Measurement Methods to Determine Compliance with Ambient Air Quality Standards for Suspended Particles” Journal of Air and Waste Management Association, 45, 320-382, 1995 [53] R.M. Harrison, M. Jones and G. Collins “Measurements of the Physical Properties of Particles in the Urban Atmosphere”, Atmospheric Environment, 33, 309-321, 1999 [54] G. Ayers, M.D. Keywood and J.L. Gras “TEOM vs. Manual Gravimetric Methods for Determination of PM2.5 Aerosol Mass Concentrations” Atmospheric Environment, 33, 3717-3721, 1999 Volume 2
Low Levels of Particulates
Page 81
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
[55] L.F. Salter and B. Parsons “Field trials of the TEOM and Partisol for PM10 monitoring in the St Austell china clay area, Cornwall, UK” Atmospheric Environment, 33, 2111-2114, 1999 [56] A. Soutar, M. Watt, J.W. Cherrie and A. Seaton “Comparison between a personal PM 10 sampling head and the tapered element oscillating microbalance (TEOM) system” Atmospheric Environment, 33, 4373-4377, 1999 [57] J. Cyrys, G. Dietrich, W. Kreyling, T. Tuch, and J. Heinrich “PM2.5 Measurements in Ambient Aerosol: Comparison between Harvard Impactor (HI) and the Tapered Element Oscillating Microbalance (TEOM) System” The Science of the Total Environment, 278, 191-197, 2001 [58] B.M. Meyer, H. Patashnick, J.L. Ambs, and E. Rupprecht “Development of a Sample Equilibration System for the TEOM Continuous PM Monitor” Journal of Air and Waste Management Association, 50, 1345 –1349, 2000 [59] W. Wang, X. Sun, R. Bata, M. Gautam, N. Clark, G. M. Palmer, and D. Lyons “A Study of Emissions from CNG and Diesel Fueled Heavy-Duty Vehicles” SAE Paper 932826, Detroit USA 1993 [60] S. Unnasch, D. Lowell, F. Lonyai, L. Dunlap and C. Sullivan
“Performance and emissions of clean fuels in transit buses with Cummins L10 engines” SAE paper 931782, Detroit USA 1993 [61] Code of Federal Regulations, Title 40, Part 86, Subpart N, Sections describe the methodology used in Certification Tests of Heavy-Duty Engines [62] C.A. Sharp, T.L. Ullman, and K. R. Stamper “Transient Emissions from two Natural Gas-Fueled Heavy-Duty Engines SAE Paper 932819, Detroit USA 1993 [63] S. G. Fritz, and R. I. Egbuonu “Emissions from Heavy-Duty Trucks Converted to Compressed Natural Gas SAE Paper 932950, Detroit USA 1993 [64] http://www.cseindia.org/html/cmp/air/myths_facts/myth3.htm [65] M.J. Bradley & Associates, and West Virginia University Heavy Duty Transportable Emissions lab. Internal Brief: “Comparison of Emissions Performance for Alternative Fueled (CNG) and Conventional Fueled Heavy-Duty Transit Vehicles” June 1999 [66] Harvard Center for Risk analysis “Fueling Heavy Duty Trucks: Diesel or natural Gas” Risk in Perspective 8:1, 2000 Volume 2
Low Levels of Particulates
Page 82
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
[67] Z.D. Ristovski, L. Morawska, J. Hitchings, S. Thomas, C. Greenaway, and D. Gilbert Particle Emissions from Compressed Natural Gas engines” Journal of Aerosol Science, 31, 4, 403-413, 2000 [68]U.S. Department of Energy “Natural Gas Buses: Separating Myth from Fact” Clean Cities Fact Sheet series, Report number NREL/FS-540-28377, Washington, DC [69] C. S. Weaver, and M. Balam, “Comparison of In-Use Emissions from Diesel and Natural Gas Trucks and Buses” SAE Paper 2000-01-3473, Detroit USA 2000 [70] Jack Peckham “CNG Cancer Risk Worse than Clean-Diesel with PM Trap: Swedish Study Diesel Fuel News, Issue: July 14, 2000 [71] Jack Peckham “Toxicologist “appealed” at ignoring CNG Risk” Diesel Fuel News, Issue: April 16, 2001
Volume 2
Low Levels of Particulates
Page 83
ANNEX to VOLUME 2 Partial Flow Dilution Systems
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
0 CONTENTS 0 Contents 86 1 Horiba Partial Flow Dilution Tunnel MDLT-1300T Series 87 1.1 Applications.............................................................................................................................87 1.2 Benefits....................................................................................................................................87 1.3 Principles of operation.............................................................................................................87 2 AVL SPC 472 Smart Sampler 89 2.1 Applications.............................................................................................................................89 2.2 Benefits....................................................................................................................................89 2.3 Technical Features...................................................................................................................90 2.4 Technical Data.........................................................................................................................90 3 Control System PSS-20 91 3.1 Benefits....................................................................................................................................91 3.2 Dilution air system ..................................................................................................................91 3.3 Diluted gas system...................................................................................................................91 3.4 Mobile sampling system..........................................................................................................92 3.5 Control unit .............................................................................................................................92 3.6 Technical Features...................................................................................................................92 4 Model BG-1 Micro-Dilution Test Stand 93 4.1 OPERATION ..........................................................................................................................93 4.2 Sampling..................................................................................................................................93 4.3 Standby....................................................................................................................................93 4.4 Remote Operation....................................................................................................................94 5 Transient Minitunnel MICROTROL 95
Volume 2
Low Levels of Particulates
Page 86
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
1 HORIBA PARTIAL FLOW DILUTION TUNNEL MDLT-1300T SERIES The Partial Flow Dilution Tunnel MDLT-1300T series is a proportional sampler that measures diesel particulate emissions using the partial flow dilution method. Designed for transient engine testing, the system employs innovative and fast-acting flow rate control to keep the exhaust volume that enters the tunnel proportional to the total engine exhaust flow rate. Compact and versatile, the MDLT-1300T series can be used as a standalone system in place of a full dilution tunnel or in combination with a conventional CVS system. Applications range from transient testing of heavy-duty diesel engines to general purpose steady state testing of both large and small engines.
1.1APPLICATIONS
Particulate measurements for engine R&D and certification
Transient and steady state testing
EURO III (ESC) for heavy duty diesel engines
ECE-R49, EEC 91/542, and ISO 8178
1.2BENEFITS
Operates as a standalone system or in conjunction with a conventional CVS
Replaces full-flow dilution tunnels with excellent correlation
Compact size for small test cells
Can be transported easily from one test cell to another
Includes automatic calibration for ease of use
1.3PRINCIPLES OF OPERATION The MDLT-1300T series takes a small portion of the total exhaust, dilutes it with air to create a constant flow rate, and passes the diluted exhaust sample through filters to measure particulate matter. Highly accurate venturi flow meters control and measure the diluted exhaust flow rate and the dilution air flow rate, thus determining the partial exhaust flow (partial flow=diluted exhaust flow - dilution air flow). Two modes of operation--proportional sampling and constant dilution ratio--provide application versatility. Proportional sampling allows the system to replace full dilution tunnels for transient testing. The partial exhaust sample
Volume 2
Low Levels of Particulates
Page 87
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
that enters the tunnel is proportional to the total exhaust flow rate. Since transient test conditions cause frequent changes in the exhaust flow rate, fast-acting control of the dilution air flow is essential. The MDLT1300T series meets this challenge with an innovative design.
A CFO delivers the minimum air flow required during a test.
A piezo valve quickly adjusts this flow to the specific rate required to maintain proportionality.
The constant sampling mode allows the MDLT-1300T series to be used with a conventional full flow CVS for transient testing or as a standalone unit for steady state tests. The system can simulate both double dilution and single dilution operation when used with a CVS. The measurement configuration is shown in Figure 1.
Figure 1. Schematic of PM measurement with MDLT 1300T
Volume 2
Low Levels of Particulates
Page 88
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
2 AVL SPC 472 SMART SAMPLER The SPC 472 is a partial-flow tunnel for gravimetric measurement of diluted particulates. From the total engine exhaust flow only a small partial flow is sampled into the mini dilution tunnel and diluted with air which the system conditions internally. The system can therefore simply be operated with a factory shop air supply. The dilution ratio is adjusted and the partial flow rate set by the mass flow controllers for the dilution air and the total tunnel flow (partial flow=total tunnel flow minus dilution air flow). This principle allows not only the CVS simulation as required by the EEC 91/542 (i.e. constant total tunnel flow and constant partial flow) but also the adjustment of constant dilution ratios at constant total tunnel flow (constant mode). The operating principle of the SPC 472 Smart Sampler makes it very simple to install the welded sampling probe in the exhaust pipe on the one hand and on the other, guarantees quick response time of the mass flow controllers giving the system transient capability. Thanks to its compact design and roller wheels, the AVL 472 Smart Sampler can be used on different test beds and is quickly ready for use – for measurements on engines of all sizes in any location, from passenger car to ship engines. Synchronisation with the test bed host is via a serial interface and communication is based on the AK generic communication interface
2.1APPLICATIONS The AVL 472 Smart Sampler is a cost-efficient, high-precision partial-flow dilution tunnel for R&D applications and engine certification, satisfying the international legal requirements of ECE-R49(88/77/EEC), 91/542/EEC and ISO 8178 on particulate measurement. The system is also designed to meet the future requirements of EURO III. A standard 13 mode test is preprogrammed and user-specific measuring procedures can also be easily programmed on the system control PC – for both steady-state and transient testing.
2.2BENEFITS •
Steady-state and transient application
•
Compact design for easy transport to different engines (or test cells), efficient usage
•
One system can be used for both small and large engine sizes
•
Easy installation and operation
•
Proven correlation to the full-flow CVS tunnel
•
Sophisticated technology, the SPC 472 is the third Smart Sampler generation
•
Optional automatically switchable filter holders
Volume 2
Low Levels of Particulates
Page 89
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
2.3TECHNICAL FEATURES •
Modular design with separated system configuration (components with castors)–PC operation with familiar user interface
•
Fast flow control and temp. conditioning allow accurate measurement with high repeatability
•
Automatic calibration with inbuilt laminar flow element device
•
Easy and simple installation of the sampling probe
•
Same probe size for different engine sizes
2.4TECHNICAL DATA Power supply Control rack: 3 x 320 ...440 ± 10% VAC / 50/60 Hz, 3 kVA PC: 110/230 V / 50/60 Hz Pressurized air requirement: 6..10 bar, 200 Nl/min Weights Control rack: approx. 250 kg, PC: approx. 24 kg Dimensions W x D x H: 730 x 700 x 1640 mm Mobile dilution and particulate sampling unit space required: W x D x H: 750 x 560 x 1530 mm (with micro dilution tunnel, filter holder, by-pass and switch valves mounted on the panel) System configuration •
1 control cabinet incl. sampling pump (movable)
•
1 micro-tunnel mounted on a movable panel
•
1 PC with monitor and keyboard
Options •
Option 2 Filter Holder
•
Option 3 Filter Holder
•
Option Automatic Filter Changer
•
Option Dilution Air Cooler
•
Option Preparation For Gas Analysis
•
AK Software
•
Transient Software
Accessories •
Particle Filter
•
By-Pass Filter
•
Micro Balance
•
Clean Work Bench
Volume 2
Low Levels of Particulates
Page 90
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
3 CONTROL SYSTEM PSS-20 PSS-20 represents a low cost and user-friendly system complying with annex V of this directive. The mini dilution tunnel is the result of a joint cooperation between CONTROL SYSTEM, a company engaged in engine testing, and IVECO, started in 1987. It consists of a cabinet containing the dilution air system, the diluted gas system and the control unit, and a mobile trolley with the sampling system to be placed close to the engine.
3.1BENEFITS
User friendly,
No compressed air or water is requested,
No restriction on the running operations and on the type of engine,
It is not limited to a range of exhaust gas velocities,
Mechanical flowmeters with optical encoders to have the maximum precision on the flow measure,
Pressure control valve and by-pass system allow to start every test step in the correct conditions,
It is possible to select "volume" or "time" test sequences,
The unit is easily transportable from one test location to another,
Requires a single phase 220 V supply,
Easy handling of the filter holders.
3.2DILUTION AIR SYSTEM
inlet air filter (0.1 mm efficiency)
positive displacement flowmeter with electronic pulse generator
variable speed volumetric pump
3.3DILUTED GAS SYSTEM
heat exchangers
Volume 2
Low Levels of Particulates
Page 91
EC-DG ENTR
Emission control technology for heavy-duty vehicles
variable speed volumetric pump
positive displacement flowmeter with electronic pulse generator
ETD/00/503430
3.4MOBILE SAMPLING SYSTEM
38 mm ID stainless steel mini tunnel
two 70 mm stainless steel filter holders
filter by-pass able to maintain a constant flow of diluted gas in sampling stand-by operations of 13modes procedure
3.5CONTROL UNIT
based on 80486 PC with LCD display and integral keyboard controlling all the functions and displaying the system status and results
The system is controlled by a 80486 PC with HD 100 Mb storage and by software working in Microsoft Windows operating system. The software controls all the test bed functions, and enables you to define and to execute any sampling cycle you need with an automatic computation of time and volumes, using engine and cycle configuration files. The LCD displays individual readings of pressures, temperatures and volumes according to the EEC directives.
3.6TECHNICAL FEATURES Dimensions : Control cabinet :600 x 800 x 1800 mm (W x D x H) Mini tunnel trolley : 1200 x 300 x 800 (W x D x H) Weight : Control cabinet : 250 kg Mini tunnel trolley : 70 kg Recorder output : 0 - 10 V for all pressure and temperature probes Power supply : 220 V, 50Hz, 2 kW Dilution air flow : 0 - 20 m 3 Diluted gas flow : 0 - 20 m 3 Volume reading : 0 - 10,000 liters (resolution : 0.1 L) Control unit : 80486 PC ; HD 100Mb ; FD 3"1/2 ; 9"LCD display ; flat keyboard Communication : RS 232C Serial line for external computer Control
Volume 2
Low Levels of Particulates
Page 92
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
4 MODEL BG-1 MICRO-DILUTION TEST STAND
4.1OPERATION The BG-1 software was designed to allow operators who are unfamiliar with diesel particulate sampling to quickly begin using the system to produce results. The software consists of one main screen (figure 2) which is a schematic diagram of the hardware and flow paths of the BG-1. All major functions, including sampling, diagnostics, and calibration, are initiated from this main screen.
4.2SAMPLING Sampling using the BG-1 mainly consists of setting and accurately controlling the flow rates of the two mass flow controllers and the time required for sampling. The features of the BG-1 allow this to be easily accomplished so that the operator can concentrate on other aspects of the test procedures. The sample mode is designed to accept inputs from either a data file or from the keyboard. Both options have the same input requirements and sampling ability. The data file input allows the user to prepare a complete test consisting of any number of filter samples prior to initiating the test sequence. In both modes, keyboard and data file, all sample information is recorded to a user selectable output file on the BG-1 computer. To collect a sample using the BG-1, the operator enters all of the necessary sample parameters, as shown in figure 3, into the system. The important parameters are the total sample flow rate, dilution flow rate, or dilution ratio, and the sample time. One additional sample parameter is the stand-by time for the sample. The stand-by feature performs three important functions. First it acts to clean the sample probe of material collected on the sample probe between samples and thus reduce the potential error in the sample. The second use of the stand-by mode is to cool the dilution chamber and thus reduce thermal changes that would otherwise occur during the sample. The third use of the stand-by mode is to allow both the dilution flow and total sample flow to reach their set point values prior to initiating the actual sampling of exhaust particulate. Because of the incorporation of a standby mode, both multiple filter sampling, as defined by the EPA, and single filter sampling, as designated by the EEC, are possible.
4.3STANDBY One important mode in the BG-1 is the standby mode which is automatically initiated when collecting samples using the BG-1 in a stand alone mode. This standby mode serves to stabilize the flows through the two flow controllers prior to collecting the sample, thus increasing the accuracy of the sample. During the standby mode both flow controllers are initialized to the setpoint to be used during the sample, but instead of the system collecting a sample, the total flow is stabilized by pulling air through the standby valve, indicated as valve number 2 on the main screen. The purging function is accomplished by delivering dilution air
Volume 2
Low Levels of Particulates
Page 93
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
through the dilute flow meter, into the dilution chamber and out into the exhaust through the sample probe. The flow rate used during this mode is the same as selected by the operator for the actual sample. Enter the length of time desired for standby prior to sampling, this will typically be one minute, but can be any length desired. If the time selected is too short, the flow controllers will not have sufficient time to stabilize prior to the beginning of the sample and the results will not be as accurate as possible. If you are ready to begin sampling before the end of the standby period, briefly press the stop button on the main BG-1 screen and the sample will begin. Entering a long standby time allows you to purge for an extended period of time, such as when changing engine operating conditions, without interrupting the sample process. Calibration The BG-1 has built-in calibration features that allow the operator to easily perform many calibration functions, such as those needed to meet reporting requirements for engine certification. The main technique of calibration is know as mirroring, where the slave controller, the total mass flow controller in this case, is calibrated in reference to a master, the dilution flow controller. This calibration feature of the BG-1 allows a relative calibration of the two controllers to be performed to within 0.3% of reading over the range of operation of the system (figure 4). In addition, the system has been designed, both with the hardware and software, to allow the operator to easily connect a transfer standard to the BG-1 to perform a periodic check of the accuracy, or a recalibration, of the flow controllers in reference to the standard.
4.4REMOTE OPERATION The BG-1 remote control software that has been developed by Sierra provides the ability operate the system from a remote, or master, computer following standard communications protocols and device specific commands. This allows the customer to incorporate these commands into their existing test cell environment and control the BG-1 in conjunction with other equipment in the test cell. The communications can be accomplished with any computer, typically be the main test cell computer, communicating over the industry standard RS-232 communications protocol. This allows real time control of the BG-1 for both steady-state and transient cycle testing. Along with controlling the BG-1 from the host computer, the system is designed to provide the necessary output data to the host to facilitate coordinating data acquisition from all test cell instrumentation. The design of the BG-1 thus can thus be easily used as both a stand-alone instrument as well as integrated fully into the test cell environment. Configuring the Two Computers for Remote Operation Hardware Connections The communications between the master and slave computers is performed over a RS-232 serial link using a fully shielded LapLink serial cable. This cable is connected to the BG-1 at the serial connector on the front of the cabinet. The system uses hardware handshaking between the two computers. The communications uses standard RS-232 protocol.
Volume 2
Low Levels of Particulates
Page 94
EC-DG ENTR
Emission control technology for heavy-duty vehicles
ETD/00/503430
5 TRANSIENT MINITUNNEL MICROTROL The MICROTROL is a partial flow sampling system for the measurement of particulates in diesel engine exhaust gases.
Fully certified by TUV the Microtrol is suitable for the following directives: On Highway: •
R49, EURO II EEC/91/542 (old directive 88/77/EEC)
•
EURO III EEC/99/96, ESC (European Steady State Cycle)
•
US EPA CFR 40 PART 89
•
JAP Cycles
Off Highway: •
ISO 8178 All Variations
The Microtrol 4 fulfills the requirement for transient testing using the ETC. Making it suitable for all parts of EURO III and the proposed EURO IV. Fully automatic the Microtrol 4 presents a compact sampler and control unit suitable for direct mouning in constricted areas such as small engine test cells.
Volume 2
Low Levels of Particulates
Page 95
EC-DG ENTR
Emission control technology for heavy-duty vehicles
flow diagram:
ETD/00/503430
1. air filter 2. filter element 3. heat exchanger 4. control valve 5. pressure transmitter 6. calibration and measuring unit 7. temperature sensor 8. ball valve 9. temperature sensor 10. probe 11. pressure transmitter 12. temperature sensor 13. mess flowmeter 14. sampling pump 15. ball valve
16. command centre
Volume 2
Low Levels of Particulates
Page 96
