Emission measurements on diesel vehicles with Diesel Particulate Filters, carried out with a second generation smoke meter.

Transcription

1 30 december 2015 TEN Report Emission measurements on diesel vehicles with Diesel Particulate Filters, carried out with a second generation smoke meter. M.P. de Goede L.C.C. Andriessen Test Equipment Nederland B.V. Rijksstraatweg JD Baambrugge The Netherlands All rights reserved. Nothing from this publication may be copied, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior permission of the Test Equipment Netherlands BV.

2 Contents 1 Introduction Objective Soot particles from diesel vehicles The NEDC test The free acceleration test The smoke meter Measurement mode of the smoke meter The test setup and test in the laboratory The measurement results of the laboratory test The test setup and test at the DPF manufacturer The measurement results of the test at the manufacturer Test at the public cleansing service of the city of Almere The results of the test at the public cleansing service of the city of Almere Conclusions Recommendations

3 1 Introduction In the past few months Test Equipment Netherlands BV performed a study with the smoke meter TEN LPA : - in a laboratory in the United Kingdom. - at a manufacturer of Diesel Particulate Filters (DPF) in Belgium - at the public cleaning service of the city of Almere. The study should demonstrate that for modern diesel vehicles equipped with a DPF, a second generation smoke meter is an excellent instrument to measure the emission of soot particles. Since the laboratory and the manufacturer have all test facilities and Test Equipment BV Netherlands itself is not able to produce or modify DPF's, it has been decided to carry out the tests at these locations. The city of Almere has many trucks from various emission classes. The laboratory is equipped with quality appliances including an NEDC (New European Drive Cycle) configuration with chassis dynamometer and software, a Condensation Particle Counter (Horiba) and Particulate weight gauges. During the investigation in Belgium, Test Equipment Netherlands received full cooperation and assistance of the manufacturer who has several test DPFs. For this of course our heartfelt thanks. 2 Objective For several years, modern diesel vehicles are equipped with an oxidation catalyst, a DPF (Diesel Particulate Filter) and possibly an SRC (Selective Reduction Catalyst) to reduce harmful emissions from the vehicle. For our study is in particular the operation of the DPF is important. The pores of the ceramic filter of an average DPF, are approximately 18 nm in size. In theory, this means that all the solid particles having a larger diameter than 18 nm are not be allowed to pass. As long as the filter of the DPF is in good condition, the emission of carbon particles will be very low. However, what happens if the quality of the filter decreases, the filter becomes damaged or the DPF eventually is manipulated? How can one perceive this? Since the beginning of 2014 Test Equipment Netherlands is in the possession of a so-called second generation smoke meter, the LPA. The LPA has a compact measuring chamber, advanced electronics and measures with high-resolution. Test Equipment Netherlands is convinced that they can perceive with the LPA that the DPF is defective or manipulated, even when measuring very low soot concentrations. With a defective DPF, all kind of solid particles having a larger diameter of approximately 18 nm are allowed to pass. This emission is a collection of particles with various diameters, from small to large. The second generation smoke meter is no Particle Counter, but can certainly measure low soot concentrations. This study, consisted of various tests, must prove that a second generation smoke meter is able to measure low concentrations of soot particles and thus can detect a faulty DPFs. 3

4 3 Soot particles from diesel vehicles Initially it was thought that the emission of particles with a diameter of 10 microns (PM10) was the main cause for health problems. Recent studies show that the very small particles are responsible for these problems, particles smaller than 2.5 microns (PM2.5). We can divide the particles of diesel emissions in three modes, depending on the size of the particles: - the nucleation mode - the accumulation mode - and the coarse mode Particles in the nucleation mode are the smallest particles, these nanoparticles have a size of up to 50 nm. The accumulation mode includes all of the particles between 50 nm and 1 um. Particles between 50 and 100 nm are known as the so-called "ultrafines. The coarse mode is composed of particles larger than 1 um. In the figure below we see both the number and weight of the present particles. There is clearly seen that the particles in the accumulation mode, determine the bulk of the weight (red line), the number of particles (blue line) in the accumulation mode is significantly lower than the number of particles in the nucleation mode. Particle number, or the amount of particles is not the most important variable that we look for. Figure 1: The distribution of the particles according to size and weight (Kittelson, 1998). 4

5 4 The NEDC test The New European Driving Cycle (NEDC) is the unofficial name for a standardized test drive that is used for type approval of vehicles. It is designed to mimic the conditions on the roads in a reproducible way. On the basis of this test the fuel consumption and emissions of the vehicle are measured. The test drive consists of a four times repeated urban drive cycles and an extra urban drive cycle. In each urban driving cycle one drives 1 km in exactly 200 seconds and accelerates and stops three times. Maximum speed is 15, 32 and 50 km/h. In the extra urban driving cycle one drives 7 km in 400 seconds, with a speed of 70, 50, 70, 100 and finally 120 km/h. The overall driving cycle is 11 km long and takes 19 min 40 s. Theoretically, the test may take place on the road, but in practice, all manufacturers choose for the laboratory test. The test was previously performed on the roller bench, but today a chassis dynamometer is used. Herewith the rolling and air resistance is simulated by the dynamometer. The NEDC is also used to determine the energy consumption and operating range of electric and hybrid cars. Figure 2. The NEDC test graphically displayed. During the test with the smoke meter, use was made of a laboratory which is equipped with NEDC equipment. Interesting for the NEDC test the Horiba MEXA-2000 SPCS system. Figure 3. Diluter tunnel of the Horiba 5

6 Figure 4. Condensation particulate counter, also equipped with paper filters for a weight measurement. The Horiba system is an advanced system. The duliter tunnel dilutes the gas flow with clean ambient air and ensures that the gas flow is continuous. Thereafter, the exhaust gas flows to the Condensation Particle Counter. After the conditioning of the gas, here the measurement of the solid particles number takes place. The Horiba can measure particles from 23 nm (>50%) and expresses the quantity of particles per cm 3. Besides the measurement with the CPC, also a portion of the exhaust gas is passed through the paper filters. Prior to the test, the filters are conditioned and weighed. After the NEDC test the particles are stored in the filters and the filters are conditioned and weighed again. On the basis of the weight measurement, the amount of soot in g/km can be calculated. 5 The free acceleration test During the free acceleration test, the engine speed increases directly and without interruption from idle up to the maximum governed speed. During the test, the vehicle is not under load of a dynamometer. For the acceleration of the mass inertia of the engine, more power is needed which results in a shortage of air and thereby generates a maximum soot emission. The soot emission is measured with a partial flow smoke meter. During the test, one may choose to also measure the engine speed and acceleration time. In this way a synchronized test takes place in which also manipulation is reduced. Figure 5. Example of a free acceleration test. 6

7 6 The smoke meter For the study, tests are carried out with a new type of smoke meter, a so-called second generation smoke meter. It s the TEN LPA smoke meter. This smoke meter has compact dimensions, uses the latest technology such as new LED technology and active extraction and measures with high resolution the k-value. The display resolution is m -1, the measuring range of m -1 is composed of about 2000 discrete steps. The compact smoke chamber construction is very stable and is continuously heated to 75 C. The exhaust gas is sampled at a frequency of 50Hz. The smoke meter itself does not have a display and therefore must be connected to a gas analyser or computer. After switching on, the smoke meter needs a few minutes to warm up, so that a stable and accurate measurement can take place. Figure 6. Principle of the smoke meter. During the test in the laboratory the smoke meter is connected to the sample line of the Horiba system. The sample probe of the smoke meter is placed directly in the outgoing conduit of the exhaust. Normally, the sample probe is placed directly into the exhaust so that there is plenty of exhaust gas flowing to the smoke meter. The smoke meter has no pump. Figure 7. The smoke meter in the test setup. 7

8 7 Measurement mode of the smoke meter The existing and second generation smoke meters comply with the current, actual legislation. Under this legislation, the smoke meter must be able to measure in two different measurement modes, namely Mode A and Mode B. Mode A means that the smoke meter does not use a software based low-pass filter. In Mode B the smoke meter does make use of the low-pass filter. At this moment, all measurements are carried out in Mode B. Mode B includes a first order low-pass filter with a time constant of 435 ms, so that the response of a step function in exactly one second rises to 90%. In other words, the measured smoke value is considerably filtered and thus much slower. The maximum measured value in Mode B is always lower than in Mode A. Because during the test we carry out measurements on vehicles with DPF and therefore soot concentrations may be very low, it is not desirable to unnecessarily delaying the measurement signal. In order not to lose unnecessary measurement information, we measure during the test in Mode A, the unfiltered fast measurement mode. Figure 8. Differences Mode A and B during a test with the same smoke peak In the above figure we clearly see the difference between Mode A and Mode B. It concerns a measurement with the same soot peaks. In Mode A we measure a maximum emission peak of m -1. This is the realistic emission that the smoke meter measures in the smoke chamber. In contrast, in Mode B a maximum emission of 0078 is measured. The measured smoke value is filtered with aid of the first-order low-pass filter. In this case it means that the maximum value of fast, short smoke peak measured in Mode B, is approximately 47% lower than the value measured in Mode A. Mode A shows the realistic maximum smoke peak. As we try to detect faulty PDFs, sometimes with a very low soot concentration, during the tests we use Mode A for the detection of the maximum smoke concentration. 8

9 8 The test setup and test in the laboratory In the laboratory, two types of tests can be performed: - the NEDC test - the free acceleration test - For the test, use was made of a Ford Focus 1.6 TDCI. The Ford belongs to emission class Euro 5 and is equipped with an oxidation catalyst and DPF in one single housing. A total of three runs are performed for the test: - Ford Focus with an original fit DPF - Ford Focus with an empty DPF - Ford Focus with a damaged DPF For the NEDC tests, the Ford Focus is positioned on the chassis dynamometer in the laboratory. During the NEDC test all emissions are extracted and analyzed by Horiba measuring system. Parallel to the Horiba analyzer is the TEN LPA smoke meter connected. Figure 9. De Ford Focus 1.6 TDCI during the NEDC test. After each NEDC test, a series of free accelerations will be conducted. Each series consists of three free acceleration. Between each acceleration we wait eight seconds. After each measurement session the engine is allowed to cool, the used DPF is dismantled and replaced by another DPF. As already mentioned the first DPF is in perfect condition (original fit). The DPF for the second session is actually a fake DPF. The outside of the DPF looks identical, but the inside of the DPF is completely removed. The DPF for the third measurement session is qualified as damaged (not completely coated ceramic filter). To what extent and what damages were present, it is not easy to figure out. The measurements will show how much this PDF is damaged. 9

10 Figure 10. From left to right: original DPF, the empty DPF and damaged DPF. 9 The measurement results of the laboratory test General. During the execution of the NEDC test it is soon clear that this test is not suitable for the determination of the maximum soot emission. During the test, there is very limited acceleration and driving with a relatively low speed. The end result is an average value of a relatively quiet driving cycle. Expressed in terms of the number of soot particles per kilometer of milligrams carbon per kilometer. The cumulative particle meter shows very low to astronomical numbers, which are often difficult to interpret. With quotients as the average number of particles per kilometer, or the average number of particles per liter as a result. By contrast, the paper filters show a recognizable image. From white to completely black. The weight measurement shows the actual amount of soot. Even though this is a cumulative average, it is very recognizable and represents the total amount of soot, regardless of the size of the emitted particles. 10

11 Figure 11. The paper filters of the NEDC test. White, black and light gray. We do not go deeper into the weighted amount of soot. These values are not representative for the maximum soot emission of the vehicle. During the NEDC test the Horiba CPC shows the actual value, expressed as the amount of particles per cm 3. During the test with the original fit DPF that's only a few tens of particles per cm 3. During the test with the empty DPF the actual value is very high and fluctuates continuously. Figure 12. Actual particle values of the Horiba: particles per cm 3. 11

12 After each NEDC test, the free acceleration test takes place with the TEN LPA. There are three accelerations performed for each DPF. Below the results are shown. Free acceleration test with the original DPF Figure 13. Free acceleration test with the original DPF. The DPF is an original filter and in good condition, similar to that of a vehicle that is a few months old. During the test it's immediately clear that the soot emissions are minimal. The measured value is really just m -1. During the NEDC test this was already clear by looking at the actual measured values. The filter functions well and leaves very little soot particles through. Certainly, the number of particles greater than about 20 nm will be very small. So small that the TEN LPA measures zero. And that's not strange, there is nothing to measure, not with the Horiba and not with the LPA. 12

13 Free acceleration with the empty DPF Figure 14. Free acceleration test with the empty DPF. During this test it gets really clear what an efficient DPF does. Without the filter, the emissions are enormous. We measure values up to the top of the 2.0 m-1. This DPF looks like an original DPF, but is on the inside completely empty. All the soot particles, ranging from a few nanometers to a few micrometers, are emitted. Obviously, the smoke meter is fully capable to measure the emissions. Quickly and accurately the soot peaks are measured. In practice this could be a vehicle with a fake DPF or with a completely broken ceramic filter. Depending on the type of engine, the emissions will be high to very high. It is obvious that the filter is not working and the vehicle exceeds the emission limits. Figure 15. The empty or fake DPF. 13

14 Free acceleration test with the damaged DPF Figure 16. Free acceleration test with the damaged DPF. This DPF is classified as damaged and would not contain a fully coated ceramic filter. To what extent the damaged filter works is not easy to figure out. The measurements clearly show that the filter does not function as well as the original filter. Although the measured values are low, the smoke meter is perfectly able to detect the soot peaks. The maximum measured smoke value is m -1. It is not clear what the size is of the emitted particles. Because the ceramic of the filter itself should be still intact, it is likely that the emitted particles are very small. Apart from the particle size, in any case the number of particles must be fairly increased. For the LPA smoke meter in any event, enough to be able to be measured. Figure 17. NEDC test paper filter. 14

15 In the above figure, the filter paper of the NEDC test for this DPF is shown. Although the emissions are low, we have actually measured the emissions with the smoke meter. Irrespective of the exact amount of soot on the filter, it is clearly observable that a emission is measured. The white filter is changed in light gray. A summary of the test in the laboratory is shown in the figure below. Figure 18. Overview of the measured values during the smoke test in the laboratory. The values vary from 0 to more than 2 m -1. All intermediate values can be measured well with the LPA smoke meter. Low values such as m -1 are excellent perceptible. 15

16 10 The test setup and test at the DPF manufacturer For the investigation Test Equipment Netherlands have also carried out tests at a DPF manufacturer in Belgium. Also the manufacturer possesses high quality equipment in order to develop and test its products. In addition to the equipment, the manufacturer is also equipped with various testing DPFs. This test DPFs have been manipulated in order to simulate a variety of degrees of leakage in the DPF. After the modifications, the DPFs are closed again and they can be tested on one and the same vehicle. Figure 19. Two of the manipulated DPFs. As can be seen in the figure above, on the back side of the ceramic filter there a number of drilled holes of approximately 14 mm. So in the boreholes there is a number of channels of which the rear side is fully open. Through these channels particles can flow out freely. The particles have a size from a few nanometers to a few micrometers. Depending upon the number of drill holes, there will be more particles that can flow out freely. For the test we had disposal to the following filters: DPF 1 DPF 2 DPF 3 DPF 4 DPF 5 Original fit DPF 5 holes 9 holes 17 holes 25 holes At the manufacturer only the free acceleration tests were conducted. The NEDC test is not performed since this test is not intended to measure the maximum soot emission. The free acceleration test is quick and simple to perform. Of course, even more important is that we can immediately detect whether the particulate filter works properly or too much soot particles pass the filter. During the test, we make use of an Opel with 2.0 CDTi engine. The vehicle has km on the odometer. Incidentally, this does not say anything about the condition of the DPF, as it already has been replaced several times for testing purposes. 16

17 Figure 20. The test facility at the manufacturer. For the free acceleration test sample probe of the smoke meter is placed directly in the exhaust. All measurement data is stored on the laptop. In total we perform five test runs. After each test, the vehicle is moved to the garage next to the test room to exchange the DPF. Figure 21. The LPA smoke meter during the free acceleration test. 17

18 11 The measurement results of the test at the manufacturer Free acceleration test with original DPF The following graph shows the emissions of the vehicle with original DPF. The emission was very low, but certainly evident. The maximum measured smoke peak was m -1. Figure 22. Measured smoke emission with the original DPF. Then the tests were carried out with the manipulated filters. The DPF with holes 5, 9 holes, 17 holes, and finally with 25 holes. Figure 23. Example of the DPF with 17 holes. 18

19 Free acceleration test with DPF with 5 holes Figuur 24. Measured smoke emission with DPF with 5 holes. Shown in the graph above is the emission of the vehicle with the DPF with five holes. The soot emissions has increased enormously. Again it becomes clear how important a properly functioning DPF is. The maximum measured smoke peak is no less than m -1. Figure 25. DPF with 5 holes 19

20 Free acceleration test with DPF with 9 holes Figure 26. Measured smoke emission with DPF with 5 holes Shown in the graph above is the emission of the vehicle with the DPF with nine holes. Again the soot emissions has increased. The maximum measured smoke peak is no less than m -1. Figure 27. The DPF with 9 holes. 20

21 Free acceleration test with DPF with 17 holes Figure 28. Measured smoke emission with DPF with 17 holes. The soot emission begins to take significant values. The maximum measured smoke peak is no less than m -1. A real super polluter. 21

22 Free acceleration test with DPF with 25 holes Figure 29. Measured smoke emission with DPF with 25 holes. When measuring the DPF with 25 holes, the soot emission increases further. Up to a maximum soot emissions of approximately 1.65 m -1. The smoke meter also measures these high readings without any problem. 22

23 The figure below shows all test results summarized in one single graph. The soot emission of the vehicle in combination with the manipulated filters is clearly visible. The measured smoke peaks vary between approximately 0.4 m- 1 to 1.65 m -1. The measured values provide a stable and accurate representation of the emissions. Figure 30. Results of the DPF test. The TEN LPA is perfectly capable of measuring the emissions of defective DPFs. High opacity readings and very low readings. Since there are a number of open channels in the manipulated DPFs, very small to very large particles may be emitted. During the test, it is clear to see who the polluters are. Depending on the statutory smoke limit, these filters would have failed. 23

24 12 Test at the public cleansing service of the city of Almere The public cleansing service of Almere has various trucks for collecting of household refuse. As part of a study by the city of Almere, Test Equipment Netherlands carried out emission tests on various trucks of this service. Figure 31. Trucks of the public cleansing service of Almere. The city of Almere has various vehicles from different emission classes, namely Euro III, Euro V and Euro VI. The council of Almere wanted to map out what the emissions are from different types of vehicle. The test vehicles are: - Daf, Euro III - Daf, Euro V - Daf, Euro VI An interesting detail is the fact that the Euro III vehicle is equipped with a semi-open retrofit particle filter. The Euro V and Euro VI vehicle are in original condition as DAF supplied them. The measurements were performed throughout Almere and during collecting the household refuse. The advantage is that the engines are at operating temperature and the emission is measured as realistic as possible. The trucks were put to the side of the road in order to carry out a free acceleration test. The test setup was mobile and accommodated in a vehicle with a power generator. 24

25 13 The results of the test at the public cleansing service of the city of Almere The figure below shows the measurement results of the test at the public cleansing service of Almere. One can immediately see that the particles emission of all three vehicles is low. The Euro VI is the cleanest vehicle. The LPA smoke meter measures m-1. The DPF filters out almost all particles larger than 20 nm from the exhaust gas. Figure 32. Test results of the trucks of the city of Almere. The red curve presents the results from the Euro III with retrofit particulate filter. This vehicle is older than the Euro V vehicle, but it emits far less soot than Euro V. The Euro V vehicle gives by far the most emissions. Of course, it should be noted that a vehicle with a maximum emission of m -1 is not really big polluter. This test in Almere shows that we are perfectly capable of using the LPA smoke meter to measure emissions of modern diesel vehicles. The values are certainly very low, but nevertheless the LPA shows that the values are measurable. During the test, no real polluters were detected. Since there will be only Euro VI vehicles purchased in the future and the current particulate emissions of the fleet is acceptable, there was no reason for the city of Almere to take immediately steps to reduce emissions of the fleet. 25

26 14 Conclusions During our study, we carried out various soot measurements on modern vehicles which make use of DPFs. The measurements were carried out with a second generation of smoke meter. If a vehicle has a good working DPF, the emissions of particles greater than 20 nm are very low. So low that the smoke meter measures zero or very low smoke values. The high resolution smoke meter is capable of doing this. Once the DPF from the vehicle is faulty, it will irrefutably be measured by the second generation smoke meter. The smoke meter is able to directly detect this polluters and to distinguish between good working and faulty filters. For the detection of malfunctioning PDFs of Euro 6 vehicles, the free acceleration test is the best option. Unlike several other test methods, the free acceleration test actually measures the maximum soot emissions. The free acceleration test can use the "fast pass", so there is no need to accelerate more than once. The test is fast and does not need any other auxiliary test equipment to carry out an emission test. The purchase price of the smoke meter is low. In contrast to other complex measuring devices, a smoke meter only costs a fraction of the price. Another advantage is that the smoke meter can be used for all categories of diesel vehicles, from old-fashioned polluters to sophisticated Euro 6 / VI vehicles. Due to the construction, the smoketer meter is suitable for rugged use in the workshop environment. A smoke meter is simple to calibrate periodically. By using modern smoke meters for measurements on Euro 6 / VI vehicles, we do not need legislation to be amended. 15 Recommendations For measuring the emissions of vehicles with advanced exhaust system, one can make use of second generation smoke meters. These meters are compact, robust and measure with high resolution. This allows that the smoke meter can also measure very low smoke concentrations. For the measurement of low soot concentrations on Euro 6 / VI vehicles, we use measuring mode A. This is the unfiltered measurement mode in which the smoke meter quickly and realistically measures the smoke peak. To detect real polluters among the Euro 6 / VI vehicles, the threshold (limit value) of such vehicles must go down further. A value of 1.0 or 1.5m-1 is still far too high. Euro 6 / VI vehicles with smoke emissions higher than 0.3 m-1 should be rejected, but that is a political / legal decision. The test procedure for Euro 6 / VI vehicles may possibly be extended with the EOBD test for measuring the engine speed and engine temperature. In combination with the "fast pass", the duration of the measurement is significantly reduced. For measurements in the low range, values of 0 to 0.5 m-1, it is recommended to use a second generation smoke meter. The high resolution of such smoke meters ensure that even very low concentrations can be measured accurately. The high resolution, in combination with a low threshold (limit value), makes the detection of polluters in the future much simpler. 26