Diesel Fuel Vaporizer: a Way to Reliable DPF Regeneration

SAE-NA 23-1-5 Diesel Fuel Vaporizer: a Way to Reliable DPF Regeneration Andreas Mayr, Jürgen Klement, Marco Ranalli, Stefan Schmidt ZEUNA STAERKER GmbH & Co KG, Augsburg, Germany ABSTRACT The implementation of the regeneration strategy on any operation point of the engine map represents the crucial factor for a safe and reliable DPF system. The concept developed at ZS-ArM offers a solution with low complexity and reduced cost which can be implemented to achieve the severe regeneration conditions in DPF systems. The first part of this work shows the preliminary investigations carried out on the engine test bench to achieve the targeted regeneration temperature. After the first very encouraging results, a prototype device was implemented on a test vehicle. Regeneration tests were successfully performed in real work operation in urban, extra-urban and motorway driving conditions. Finally, emissions measurements were performed with different exhaust system arrangements and in EUDC condition. Results of all the measurements are reported. The "on the road" tests have shown, that under typical city driving condition a reliable and complete regeneration can be reached. The unique requirement is that the initial gas temperature is above 25 C. Even in driving conditions with a higher exhaust gas flow as in a country road or highway driving regeneration temperatures can be reached. Tests carried out on the engine test bench show the same results. HC and CO emissions were also measured and do not appear to be critical. Furthermore, HC emissions only appear in the starting phase of the orizer. This could be easily minimized with an opportune design of the ramp of the fuel injection. In addition to the tests with the standard configuration - DOC + DPF (bare trap and coated) - a system with a coated DPF without DOC was run. Even though the total amount of Platinum was much smaller than before, the regeneration temperature could be reached. However, the emission of HC was rather high. INTRODUCTION In 2 the first DPF systems for passenger cars since the 8ies were introduced on a large scale. The regeneration strategy bases on raising the exhaust gas temperature by means of engine measures. Intake throttling and delayed injection raise the temperature at least to the light off temperature of the DOC, post injection creates hydrocarbons, which combust in the DOC until the exhaust gas reaches the ignition temperature of the soot. Although this strategy has been adopted and improved by some car manufacturers it has some drawbacks: the danger of oil dilution, over dimensioning of engine cooling system, overheating of both close coupled and underfloor catalyst with subsequent heat losses between close coupled DOC and underbody DOC/DPF. The most severe disadvantage is, however, that the engine measures can be hardly used in or close to idle condition. Furthermore the tuning of the engine measures causes a lot of efforts. The fuel orizer concept - increasing the HCconcentration with an additional unit in front of the underbody DOC outside of the engine - is similar to this solution, but avoiding the drawbacks. OBJECTIVES Goal of this investigation is to develop an external measure to release a reliable regeneration of a Diesel Particulate Filter (DPF) in the complete engine map, guaranteeing the durability of the system. In the first phase of the project several different concepts were tested. The concept of the fuel orizer, described in the next chapters, showed the highest potential to meet the project targets. TESTS Following tests were performed: vehicle tests to investigate the regeneration performance in different driving conditions emission tests to measure the secondary emissions caused by the orizer 1

MEASUREMENT SET UP The orizer injects gaseous diesel fuel upstream the underfloor DOC (Fig. 1). The orizer itself (shown in Fig. 2) includes a heating element, for example a glow plug in a small orization chamber. A pump transports the fuel from the fuel supply line. The pump is controlled by a simple control unit. km/h, the occasional maximum was 55 km/h. The driving was interrupted by many stops. Country Road / Highway Cycle; this test track combined both, a country road and a highway. The maximum speed on the country road was 1 km/h, the driving was interrupted by cross-town passages partial with traffic lights. On the highway the constant speed was 13 km/h. For regenerations in the AUDC the exhaust gas temperature was raised to some 35 C by engine measures to reach the light off temperature of the DOC. In the other cycles no engine measures were needed. After that, the orizer was turned on. For the control of the fuel dosage a target temperature of 6 C was chosen. Fig 1: Measurement set-up The injected hydrocarbons oxidize in the DOC. This increases the exhaust gas temperature up to soot regeneration temperature. The dosage of the fuel is controlled by the temperature sensor downstream the DOC. Fig 2: concept orizer VEHICLE TESTS The road tests were run on a 3l diesel engine test car. The engine ECU could be switched to a regeneration mode, where the exhaust gas temperature could be raised by engine measures. The exhaust system was formed by a pre-catalyst au under-floor catalyst and an uncoated DPF. The filter was a 5.32l (27mmx1mmx3mm) SiC substrate.three different driving cycles were chosen to tests the system in a wide range of driving conditions: VEHICLE TEST RESULTS Urban Driving (AUDC) Without any measures, the exhaust gas temperature in this cycle was ca. 2 C. With engine measures the gas temperature in front of the filter raised up to 35 to 4 C (Fig 3). 75 7 65 6 55 5 45 4 35 3 25 2 15 1 5 Vap on Engine mesures on velocity Vap off Engine mesures off 6 12 18 24 Fig 3: Regeneration in AUDC (I); soot load 5,2g/l, fuel consumption 71g 15 14 13 12 11 1 9 8 7 6 5 4 3 2 1 After activating the orizer the temperature upstream the DPF raised to 6 C and was kept stable until the orizer was switched off again. The increase of temperature was very fast and the regeneration temperature was reached within few minutes. After the test, the DPF was completely regenerated. The next regeneration tests are done in the stop and go zone (Fig4). Speed [km/h] Augsburg Urban Driving Cycle (AUDC); typical inner city driving condition with stop and go part. The averaged driving velocity in this cycle was about 2 2

75 7 Vap on 65 6 55 Engine 5 mesures 45 on 4 35 3 25 2 15 1 5 Vap off Engine mesures off velocity 6 12 18 24 Fig 4: Regeneration in AUDC (II); soot load 5,9g/l, fuel consumption 78g 15 14 13 12 11 1 9 8 7 6 5 4 3 2 1 Speed [Kmh] 1. HC emissions dependency on ourized fuel amount (m F ), T_upstream_DPF=const. 1.1 with coated filter 1.2 with uncoated filter 2. HC emissions dependency on exhaust gas temperature (m F = const.) 2.1 with coated filter 2.2 with uncoted filter 3. HC emissions in the ECE cycle only with coated filter 4. HC emissions with coated filter without DOC Country Road / Highway Driving Without any measure, the exhaust gas temperature was 25-3 C in the country road and 35-4 C on the highway (Fig. 5). In both driving conditions the regeneration temperature could be reached by using the orizer. No engine measures were needed. 75 7 65 6 55 5 45 4 35 3 25 2 15 1 5 COUNTRY ROAD 6 12 18 24 3 36 42 48 MOTORWAY velocity FIG 5: Temperatures reached with orizer in country road / highway driving EMISSION TESTS 375 35 325 3 275 25 225 2 175 15 125 1 75 5 25 For the tests a.76 l DOC (cordierite: 6.5mil/4cpsi) with 75g/ft3 PGM amount and a 4.1 l DPF (SiC: 5" x 6,1" x 1") with 3 g/ft3 PGM amount were used. The filters were loaded at a 1.9 JTD 8V engine with ca. 1-12g/l. The loading point was stationary at 12 rpm 6 Nm. The amount of soot was determined by weighing the filter after conditioning at 15 C in oven. The emission tests were run at a 1.9l TDI PD engine. The regeneration was initiated by the orizer. The dosage during these tests was kept constant. The deviation of the dosage was ca. 1 ml/min. Following tests were carried out: Speed [km/h] Fig 6: Regeneration set up: Before every regeneration the filter was heated with the exhaust gas. When the temperature downstream DPF was constant, the orizer was switched on. EMISSION TEST RESULTS HC emissions dependency on ourized fuel amount m F (T_upstream_DPF=const.) Test 1.1: coated filter The filter was heated at 12 rpm, 5Nm to 26 C. Then the orizer was turned on with 8 ml/min fuel dosage (Fig 7)., Pressure [mbar] 7 65 6 55 5 45 4 35 3 25 2 15 1 5 Emissions with different fuel dosages - coated filter - 3 6 9 12 15 18 time [s] Fig. 7: Coated DPF - emissions with different fuel dosage 7 65 6 55 5 45 4 35 3 25 2 15 1 5 Conc. O2 [Vol%],, _center _side The gas temperatures in front of the filter reached 5 to 53 C. No regeneration was observed. The HC emissions reached temporary 3 ppm. This test was repeated with a dosage of 16 ml/min. Now the temperatures in front of the filter climbed up to 6-65 3

, Pressure [mbar] C. The filter regenerated after some minutes (temperature peak downstream, CO peak, pressure drop decrease). This regeneration is zoomed in Fig. 8. 7 65 6 55 5 45 4 35 3 25 2 15 1 5 Emissions with different fuel dosages - coated filter - 3 36 42 48 54 6 time [s] Fig 8: Coated DPF - emissions with 16 ml/min fuel dosage 7 65 6 55 5 45 4 35 3 25 2 15 1 5 Conc. O2 [Vol%],, _center _side Again, a short peak of HC emissions was observed. It lasted some 5 s and reached 25 ppm. After this first 5 s the HC emissions were lower than 1 ppm. After this regeneration the procedure was repeated with the empty filter. The emissions were similar to the ones measured with the loaded filter. At the end of test 1.1 a regeneration with a dosage of 2 ml/min. Now the temperatures upstream the filter reached 7 C. The HC emission reached 3 ppm., Conc. coupled with an increase of CO emissions (2 ppm). At a dosage of 8 ml/min the HC emission were constant at 11 ppm, there was only few CO (8 ppm). 1 95 9 85 8 75 7 65 6 55 5 45 4 35 3 25 2 15 1 5 Emissions with different fuel dosages - uncoated - 3 6 9 12 15 18 21 24 27 3 33 36 39 42 45 48 51 time [s] Fig 9: Uncoated DPF - emissions with different fuel dosage HC emissions dependency on exhaust gas temperature ( m F = const.) These tests were run with coated and uncoated filters without soot. The fuel dosage was kept constant (8 ml/min), while the initial exhaust gas temperature was raised in three steps from 165-255 C. The results are shown in Fig 1 and 11. 5 Emissions vs. Gas Temperature - coated - 2 18 16 14 12 1 8 6 4 2 Conc. O2 [%],, Pressure [mbar] _center _side 1 Test 1.2 uncoated filter 4 8 The tests were repeated under the same conditions with uncoated filters. There are some severe differences in the emissions. [K], 3 2 6 4 Even in the first test with 8 ml/min fuel consumption a weak regeneration was observed. The HC emissions reached 16 ppm. They decreased immediately when the regeneration started. Simultaneously a CO peak of almost 9 ppm appeared. The regeneration was slow and incomplete. During the next test phase with a dosage of 16 ml/min the regeneration continued until all the soot in the filter was burned. Again, the regeneration was slow and no exothermal temperature raise was observed. This time the CO emissions were even higher (3 ppm). The HC emissions reached 6 ppm right after starting the orizer, they decreased during the regeneration and raised again to 1 ppm when the regeneration was finished. These test were repeated with the clean filter. At a dosage of 16 ml/min HC emissions raised to 15 ppm and then declined to 115 ppm. This decline is again 1 165 195 23 255 Gas Temperature upstream DOC [ C] Fig 1: Coated filter - emissions dependency on exhaust gas temperature There is almost no temperature increase over the DOC at 165 C upstream temperature. At 195 C there is a temperature increase of the temperature downstream the DOC, at 23 C the exothermal hat reaches the maximum. According to the temperature increase the concentration of HC and CO is decreased by raising the gas temperature upstream the DOC. The coated and the uncoated filters show qualitatively the same results. By gas temperatures over 23 C, however, the emissions of HC and CO are much smaller in case of catalytic filter - the HC disappear almost completely. 4 2

[K], 5 4 3 2 Emissions vs. Gas Temperature - uncoated - 1 8 6 4 2 ml/min 6 to 65 C were reached, with 25 ml/min fuel the temperatures upstream the DPF went up to 65 to 7 C. These temperatures showed a very strong fluctuation due to the different engine conditions in the ECE cycle. With a controlled fuel dosage, like it was used in the vehicle tests, these fluctuations would be reduced significantly. 7 Emissions And Temperatures during ECE cycle - coated / dosage 25 ml/min - 56 1 2 65 6 _center 52 48 165 195 23 255 Gas Temperature upstream DOC [ C] Fig 11: Uncoated filter HC emissions dependency on exhaust gas temperature HC Emissions vs. Gas Temperature coated / uncoated 55 5 45 4 35 3 25 2 15 _side 44 4 36 32 28 24 2 16 12 Conc. O 2 [Vol%],, 25 1 8 5 4 coated uncoated 3 6 9 2 Fig. 14: Vaporizer with dosage 25 ml/min in ECE cycle (coated filter) HC emissions [ppm] 15 1 5 165 C 195 C 235 C 255 C T_upstream_DOC Fig12: HC emission vs. Gas Temperature (coated / uncoated) In both tests there are two HC peaks during the tests. The first one is measured even before the fuel injection is started. This may be caused by fuel that remained in the or chamber. This fuel gets orized when the glow plug is heated before the fuel injection is started. It may not appear, if the orizer is heated some minutes after the fuel supply was interrupted at the previous regeneration. The second peak is observed some 3 s after the start of the fuel injection. It disappears with increasing filter temperature. HC emissions in the ECE cycle: Test 3.1 with coated filter These test were run with constant fuel dosage. In the first test the dosage was 2 ml/min, in the second 25 ml/min. The glow plug was heated ca. one minute before the fuel was injected into the our chamber. 7 65 6 55 5 45 4 35 3 25 2 15 1 5 _center _side Emissions And Temperatures during ECE cycle - coated / dosage 2 ml/min - 3 6 9 Fig 13: Vaporizer with dosage 2 ml/min in ECE cycle (coated filter) In both cases the temperature upstream the DPF raised after switching on the orizer. With a fuel dosage of 56 52 48 44 4 36 32 28 24 2 16 12 8 4 Conc. O 2 [Vol%],, The maximum HC emission was 15 ppm at 2 ml/min and 48 ppm at 25 ml/min. There are also some CO emissions (2 ppm and 4 ppm). HC emissions with coated filter without DOC test 4 65 6 55 5 45 4 35 3 25 2 15 1 5 Coated Filter w/o DOC 6 12 18 24 3 36 42 48 54 6 66 Fig. 15: Vaporizer for coated filter without DOC 65 6 55 5 45 4 35 3 25 2 15 1 5 Conc. O 2 [Vol%],, 5

This test was carried out with the coated filter only (without DOC). The PMG amount was 3 g/ft 3. So the total amount of Platinum was much lower than in the tests run with DOC. Before the test was started the gas temperature was 3 C upstream and 27 C downstream the DPF. The engine run at 12 rpm and 7 Nm. The orizer was turned on and the dosage was slowly raised to 25 ml/min. Immediately the temperature started raising. The maximum fuel dosage was reached after 5 s. Then the gas temperature downstream the DPF climbed to 57 C and the regeneration started. Immediately after turning the orizer on the HC emission reached a maximum of 2 ppm, decrease to 13 ppm when the gas temperature raised to 4 C and increased to 4 ppm when the full 25 ml/min dosage was reached. The regeneration was accompanied by a high CO emission. CONCLUSION AND OUTLOOK The diesel fuel orizer appears to be a valid alternative to engine measures for complete ad reliable filter regeneration. Utilisation with coated soot filters (CSF) without DOC will be further investigated. Possible application as exhaust gas enrichment in diesel NOx after-treatment system will represents also a target for future investigations REFERENCES 1. K. Ohno, K. Shimato, N. Taoka, H. Santae, T. Ninomiya, T. Komori and O.Salvat: Characterization of SiC DPF for Passenger Car. SAE Paper 2-1- 185 2. M. Ranalli, P. Zelenka, S. Schmidt and G. Elfinger: An Active Regeneration Aid as a Key Element for Safe Particulate Trap Use. SAE_NA Paper 21-1- 62 3. P. Kugland, E. Krieger and E. Santiago: Cleaner Diesels Full Flow Soot Filter regeneration Systems SAE paper 91133 4. G. Guo, N. Xu, P.M. Laing, R.H. Hammerle and M.Matti Maricq: Performance of a Catalyzed Diesel Particulate Filter System During Soot Accumulation and Regeneration. SAE 23-1-47 CONTACT Marco Ranalli, concepts and new technologies, Center of Competence - Emissions Marco.ranalli@arvinmeritor.com ZEUNA STAERKER GmbH & Co.KG, Biberbachstrasse 9 D-86154 Augsburg, Germany 6