A Study on the Impact of DPF Failure on Diesel Vehicles Emissions of Particulate Matter
Round 1
Reviewer 1 Report
This study deals with the identification of different DPF failures with a non-destructive method as well as the impact of these failures on diesel vehicle particle emissions and specifically on opacity and particle number concentration. DPF failures mainly influence particle number concentration and less the opacity. The results are compared with Korean smoke opacity limits at periodic technical inspection as well as with particle number limits at PTI in some European countries. The paper is well structured, the objectives are clear but in many cases the authors fail to put the results in the right context and literature review is very poor. I propose major revision before publishing. Below you can find all the specific corrections I propose.
Title
The title can be improved. My proposal is: “A study on the impact of DPF failure on particulate matter emissions of diesel vehicles”
Abstract
General comment: Smoke concentration is sometimes reported as soot emissions. It is not clear sometimes what is meant as also PN concentration maybe soot emissions. I would propose to keep it consistent in the paper and preferably report what is really measured thus, smoke (opacity).
Line 17: The DPF can be damaged but can also be manipulated. The authors could do this addition in the abstract but also in the paper especially for the case of a hollow.
Lines 20-22: What was identified was that: four types of DPF failures (crack, melting, plug, hollow) affect particle number and smoke emissions.
Line 26: No significant difference between normal and crack damaged DPF for “smoke emissions”
Lines 34-36 The last sentence of the abstract can be deleted as it is a repetition.
Introduction
General comment: The literature review of the introduction is very poor. Moreover, I believe that in order to put the paper results in the right context a more extensive discussion of the particulate measurements performed during PTI in Korea and in Europe should be done. Below you can find some specific proposals.
Lines 40-41: At least one reference should be added at the first sentence of the introduction
Lines 53-55: The effect of DPF failure on particle emissions has been studied in the following paper https://doi.org/10.1007/s40825-019-00128-z
Line 56: Exhaust gas or particle emissions?
Line 59: Please check previous works showing that smoke opacity tests cannot identify DPF malfunctions;
https://doi.org/10.3390/ijerph19137602
doi:10.4271/2019-01-1190
Lines 57-60: After reporting possible problems to identify broken DPFs, the authors could report that: in some European countries, DPF failures are detected with a newly introduced particle number measurement method at idling (check the following references: TNO 2017 R10530, https://doi.org/10.1016/j.measurement.2023.112839, https://doi.org/10.3390/s20205790, https://doi.org/10.1016/j.aeaoa.2020.100095, https://doi.org/10.3390/ijerph19137602)
Lines 60-70: These lines can become a new paragraph as they describe what was done in this paper.
Line 64: CR X-ray is reported here for the first time and the acronym CR should be explained
Experimental apparatus and procedure
General comment: The authors do not report how the damages at the DPFs were done (when driving or in the laboratory?).
Lines 75-76: Any reference for the CR X-ray imaging technique?
Lines 77-78: Any reference for the preliminary study?
Lines 89-90 and 92-94: The authors repeat the same information on the instruments used at the campaign. In my opinion the sentence “It is linked…. In real time” can be removed.
Lines 98-100: NPET is portable but not PEMS. It is designed for PN-PTI measurements and is type-examined in Switzerland. You may also check research works that used the instrument for this scope: https://doi.org/10.3390/s21248325, doi:10.4271/2019-01-1190, https://doi.org/10.1016/j.jaerosci.2023.106182)
Line 101: “catalyst remover” -> replace with more commonly used term “catalytic stripper”
Line 102: “23 to13 1000” please delete 13
Line 103: “103-5” is this a typo? The upper concentration limit is 5x106
Table 1: The legend already describes the table and the “Specification of the vehicle” sentence in the table can be removed.
Table 1: Max. load 0 kg?
Line 159: “particles per second” or “Particles per cubic centimeter”?
Lines 162-163: Many information on PN-PTI tests in European counties are missing. For example, the German PN limit is not reported (250,000). Moreover, some information on the procedure should be added like that this is an idling and not a driving test. Thus, it is different from the driving cycle of this paper. You may find relative references in my comment at the introduction
Results
Line 181: “et” is a typo?
Figure 5: i) Y-axis is currently reported as smoke concentration (%). This has to be explained in the text as a concentration cannot be (%)
ii) The legend of Figure 5 should include more precise information. The figure does not plot exhaust emission which is very general but smoke for different DPF damage cases. Furthermore, it is useful for the reader to know that the standard deviation plotted is calculated for 10 repetitions.
iii) “Crack” bar has the same color as the standard deviation
Lines 251-276: The use of the verb “confirm” in this paragraph is not appropriate. It was observed (not confirmed) that PN increased
Lines 251-276: The results reported in this paragraph (maximum and average PN) are compared to European limits set for PN-PTI. This comparison makes sense only if you compare the limits with idling PN emissions. Thus, I would propose to the authors to calculate the PN emissions for the different DPFs at idling (first 5-10 seconds of the cycle) and then perform the comparison. Moreover, a comparison also with the German limit would be interesting.
Figure 6: For normal DPF and Crack DPF the concentrations are much less than 1x107 and they cannot be distinguished in the figure. I would propose to make the y-axis logarithmic in order to show clearly the differences in the different cases.
Conclusions
After the changes in the analysis of the results (idling concentration calculation etc) also some of the conclusions may change.
Author Response
Figure 6: For normal DPF and Crack DPF the concentrations are much less than 1x107 and they ca 0nnot be distinguished in the figure. I would propose to make the y-axis logarithmic in order to show clearly the differences in the different cases.
Lines 251-276: The results reported in this paragraph (maximum and average PN) are compared to European limits set for PN-PTI. This comparison makes sense only if you compare the limits with idling PN emissions. Thus, I would propose to the authors to calculate the PN emissions for the different DPFs at idling (first 5-10 seconds of the cycle) and then perform the comparison. Moreover, a comparison also with the German limit would be interesting.
- The indicator we want to show in figure.6 is that damaged DPFs exceed the Netherlands’ PN regulation, and we want to show the trend.
- In the case of the models, the level was lower than Euro 3 in the past, but through a project conducted by the Ministry of Environment of the Republic of Korea, a retrofit DPF was installed to meet the Euro 4 level. Therefore, among the PN regulation methods of European countries, the method of the Netherlands was adopted. Germany's PN regulation method is suitable for Euro 6 level vehicles, so it seems difficult to apply this data.
- ECE and NEDC, which are diesel vehicle measurement tests conducted in Europe, are not conducted under harsh conditions. In this study, we tried to compare the PN emission pattern under severe conditions according to the DPF filter failure type. So, the experiment was conducted with the KD-147 mode test currently in progress in Republic of Korea.
- The DPF can be damaged but can also manipulated especially for the case of a hollow. Then the filtration performance is significantly reduced, so excessive amounts of smoke and ash are emitted during driving, resulting in environmental pollution.
Other grammatical, vocabulary, and reference-related points have been corrected as much as possible.
Thank you.
Reviewer 2 Report
The article is devoted to an urgent problem - the protection of the environment from soot components during the combustion of fuel in cars. Particulate filters are designed to solve this problem.
Along with this, there are several comments on the article:
1. The list of references contains a very small number of publications on the subject. It is necessary to expand it and bring later publications, there are many of them in the framework of this scientific study.
2. It is necessary to give the reliability characteristics of particulate filters in the article.
3. It is necessary to make a comparative analysis of the methods for solving the problem, not only in Korea, but also in other countries of the world.
4. In conclusion, we should emphasize and highlight the scientific significance of this work and, separately, the practical usefulness for filter manufacturers.
Author Response
- The list of references contains a very small number of publications on the subject. It is necessary to expand it and bring later publications, there are many of them in the framework of this scientific study.
- Increased the number of references from 10 to 22. In addition, the contents of the references were filled in to enrich the thesis.
- It is necessary to give the reliability characteristics of particulate filters in the article.
- “The DPF used in this study were 300 samples that were attached to the vehicle after being certified by the Korea Automobile Environment Association but were separated from the vehicle and stored separately for reasons such as ” Line 63~65 added.
- It is necessary to make a comparative analysis of the methods for solving the problem, not only in Korea, but also in other countries of the world.
- “Waseda University and National Institute of Advanced Industrial Science and Technology in Japan conducted a study on the transfer and accumulation of ash based on images taken using the CT X-ray technique. Although the CT X-ray technique can accurately photograph the inner substrate of the DPF, it was judged not suitable for commercialization of DPF damage diagnosis. Because when using CT X-ray technique, it takes long time and had to remove DPF from vehicle.” Contents added.
- In conclusion, we should emphasize and highlight the scientific significance of this work and, separately, the practical usefulness for filter manufacturers.
- In this experiment, it is difficult to obtain objectivity because each DPF damage type sample is not extracted from the same vehicle. Therefore, it is expected that more accurate results will be obtained if the experiment is conducted by randomly damaging a new DPF in the same vehicle in the future. And also it seems necessary to conduct research to derive an accurate correlation between PN emission tendency for each DPF failure types.
Round 2
Reviewer 1 Report
Even if the authors did not answer point-by-point my comments (as they should have done), I reviewed the revised manuscript. The authors have improved the manuscript but there are still some points to be answered before publication.
My main comment is that the authors still confuse driving cycles with the PN-PTI test which is performed at low idling (0 km/h and no accelerations etc). For an explanation of the procedures followed in Netherlands, Germany, and Switzerland please check the paper: https://doi.org/10.3390/ijerph19137602). Thus, any comparison of the maximum or average PN concentrations of the Korean driving cycle are irrelevant. Only the PN concentration at idling can be compared and that is the reason I asked from the authors to calculate it for the idling part of their driving cycle. In summary, I propose that Figure 6 remains in the paper because it has very interesting results but the authors if possible determine the PN emissions during the first seconds of their tests when the vehicle is idling and report these values in the text or in a dedicated table.
Minor comments
Line 57
Line 59: Please check previous works showing that smoke opacity tests cannot identify DPF malfunctions (also asked during my first review)
Line 105: NPET is portable but not PEMS. It is designed for PN-PTI measurements and is type-examined in Switzerland (see my comment also at the first review)
Lines 136-140: My comment has been misunderstood. The European test during PTI is a test at low idling and not a driving cycle. Regarding ECE and NEDC I agree with the authors and indeed these two cycles do not exist anymore as they were replaced by WLTC. Thus, in my opinion there is no reason to compare at these lines the Korean cycle with ECE and NEDC. The reference to ECE-NEDC can be removed
Lines 174-177: Which is the relevance of the two cited studies [12, 13] with PN-PTI regulation in Netherlands?
Figure 6: Specify at the legend that results are averaged over 10 repetitions
Figure 6: I believe that y-axis could be logarithmic because lower PN values are not clear.
Line 320: “In idle condition”, in the paper there is no result for idle condition. If there will be some idling PN data presented, then the authors can conclude if this value is lower than the NL limit.
Author Response
Q. My main comment is that the authors still confuse driving cycles with the PN-PTI test which is performed at low idling (0 km/h and no accelerations etc). For an explanation of the procedures followed in Netherlands, Germany, and Switzerland please check the paper: https://doi.org/10.3390/ijerph19137602). Thus, any comparison of the maximum or average PN concentrations of the Korean driving cycle are irrelevant. Only the PN concentration at idling can be compared and that is the reason I asked from the authors to calculate it for the idling part of their driving cycle. In summary, I propose that Figure 6 remains in the paper because it has very interesting results but the authors if possible determine the PN emissions during the first seconds of their tests when the vehicle is idling and report these values in the text or in a dedicated table.
A. Currently, PN and PTI are scheduled to be introduced in Korea. Therefore, the characteristics of damaged PN emission were identified in the currently implemented inspection system. In the future, as a follow-up study, we plan to conduct research on the number concentration measurement and analysis by damage type under idle conditions.
Minor comments
Q. Line 57 ~ Line 59: Please check previous works showing that smoke opacity tests cannot identify DPF malfunctions (also asked during my first review)
A. In Korea, it is considered normal DPF if it satisfies the smoke concentration standard of 10% or less. Referring to Fig.5 as verification data for this, it is possible to confirm that it satisfies the domestic smoke concentration standard regardless of the type of damage. This proves that damaged DPFs that have passed the existing smoke concentration test may be used as they are mounted on vehicles.
Q. Line 105: NPET is portable but not PEMS. It is designed for PN-PTI measurements and is type-examined in Switzerland (see my comment also at the first review)
A.
Q. Lines 136-140: My comment has been misunderstood. The European test during PTI is a test at low idling and not a driving cycle. Regarding ECE and NEDC I agree with the authors and indeed these two cycles do not exist anymore as they were replaced by WLTC. Thus, in my opinion there is no reason to compare at these lines the Korean cycle with ECE and NEDC. The reference to ECE-NEDC can be removed.
A. The KD-147 mode test shows a similar aspect to the ECE driving mode. Considering this, it was adopted as a driving test method only in this experiment.
Q. Lines 174-177: Which is the relevance of the two cited studies [12, 13] with PN-PTI regulation in Netherlands?
A. In the process of revising the thesis, it was confirmed that it was entered incorrectly. thank you
Q. Figure 6: Specify at the legend that results are averaged over 10 repetitions
A. Completed
Q. Figure 6: I believe that y-axis could be logarithmic because lower PN values are not clear.
A. Since the low values of the y-axis are difficult to see, only the indicators for Normal and Crack DPFs with low indicators were separately extracted and regraphed. The corresponding graph was added as Fig.6.
Q. Line 320: “In idle condition”, in the paper there is no result for idle condition. If there will be some idling PN data presented, then the authors can conclude if this value is lower than the NL limit.
A. I deleted the “In Idling condition”
Reviewer 2 Report
When answering the second question, you gave the total number of filters. But I asked for the characteristics of reliability.
When answering the third question, you added the results of Japanese research. It would be nice to add data for a couple more countries from Europe or America.
Author Response
Q. When answering the second question, you gave the total number of filters. But I asked for the characteristics of reliability.
A. We thought that "the characteristic of reliability" meant the following.
Retrofit DPFs have been installed in vehicles with certification from the Korea Automobile Environment Association. Among them, the damaged ones were collected and stored by the Korea Automobile Environment Association. In this experiment, 300 damaged DPFs stored through the above method were randomly selected and tested.
If we have misunderstood any part of the editor's question, we will correct it in detail.
Q. When answering the third question, you added the results of Japanese research. It would be nice to add data for a couple more countries from Europe or America.
A. This time, three more studies conducted in Europe and the United States were investigated and compared with my own study. The studies are followed below.
- Nondestructive X-ray Inspection of Thermal Damage, Soot and Ash Distributions in Diesel Particulate Filters 2009-01-0289
DOI: https://doi.org/10.4271/2009-01-0289
- Ash Permeability Determination in the Diesel Particulate Filter from UltraHigh Resolution 3D X-Ray Imaging and Image-Based Direct Numerical; Carl Justin Kamp, Shawn Zhang, Sujay Bagi, Victor Wong, Greg Monahan, Alexander Sappok, Yujun Wang SAE International Journal of Fuels and Lubricants, Vol. 10, No. 2 (June 2017), pp. 608-618 (11 pages)
- Characterizing Diesel Particulate Filter Failure During Commercial Fleet Use due to Pinholes, Melting, Cracking, and Fouling; Received: 17 August 2015 / Revised: 26 February 2016 / Accepted: 2 March 2016 / Published online: 8 April 2016 # Springer International Publishing Switzerland 2016
