Verifying the Efficiency of a Diesel Particulate Filter Using Particle Counters with Two Different Measurements in Periodic Technical Inspection of Vehicles

The article presents the possibility of verifying the efficiency of a diesel particulate filter (DPF) with the use of particle counters using two different measurement methods. The tests were carried out at a vehicle inspection station using a condensation particle counter (CPC) and a diffusion charger (DC). This article presents the results of measurements of 50 vehicles. Removal of the diesel particulate filter from a vehicle is prohibited but is a known phenomenon throughout the EU. The task of periodic technical inspections is to eliminate vehicles that are inoperative and do not meet the environmental protection requirements. However, to date, European vehicle inspection stations do not have an effective tool to counter tampering with diesel particulate filters. The performed measurements allowed us to prove the hypothesis that both methods of measurement allow the effective confirmation of the presence of DPF in a vehicle during the periodic technical inspection of the vehicle and verification of the quality of its operation. In addition, the advantages and disadvantages of both measurement methods were assessed.


Introduction
The emission of toxic substances from vehicle exhaust systems is a very serious problem. Among the emitted components, particulate matter is a particular hazard. Initially, it was only checked by mass measurement. With the development of the automotive industry and the complexity of engine systems, another problem appeared-particles with much smaller diameters. It turned out that there could be thousands of smaller particles in the same unit of mass. They are extremely dangerous because they can get into the bloodstream through the lungs and thus settle in internal organs. They are also pathogenic and carcinogenic. In the European Union, great importance is attached to environmental protection, including reducing the negative impact of transport on air quality [1]. A number of procedures are applied to ensure low emission of pollutants from vehicles, including type-approval procedures [2].
For several years now, the emission of toxic components has also been evaluated in real drive conditions [3], along with measurements of gaseous pollutants, and measurements of the mass and number of solid particles [4]. These tests have the nature and accuracy of laboratory testing. It is also a problem to comply with the high standards adopted in type-approval procedures during the operation of the vehicle [5]. The currently adopted assessment method (free acceleration method) used in non-laboratory tests is ineffective for vehicles equipped with a particulate filter. This problem is due to the widespread use in Europe of measuring devices at inspection stations, which are inadequate to verify the wear status of diesel particulate filter (DPFs). The authors focused on the requirements and problems in the EU, as it is one of the largest automotive markets with the ambition to be a leader in environmental protection. Unfortunately, vehicle users often decide to remove DPFs and mask this fact with a filter emulator [6]. Such a condition is legally unacceptable, but also difficult to diagnose with currently used measuring instruments [7].
According to the directive 2014/45/EU of the European Parliament and of the Council, vehicles should maintain the properties specified in the type approval, and finding a lack of a DPF during the periodic technical inspection should result in a negative test result [8].
Numerous studies have been carried out to solve the presented problem [9][10][11][12]. The effectiveness of the presented methods is still not sufficient during periodic technical inspections. The problem is either insufficient measurement accuracy, resulting, for example, from the diversity of solid particles structures and sizes [13], or the necessity of such an invasive intervention in the vehicle's exhaust system that it cannot be used for periodic technical inspection [12].
A number of research initiatives have been established, such as the Particle Measurement Program (PMP) or New Periodic Technical Inspection (NPTI) [14] working group, aimed to develop appropriate and effective solutions.
Due to the planned mass scale of DPF system efficiency tests during periodic technical inspections, the devices used must clearly indicate whether the vehicle has passed the test or not, in terms of the filter efficiency assessment [15]. The results should not require additional interpretation.
The most advanced, specialized devices are found in laboratories conducting approval tests. However, they are very expensive and highly specialized, and are not mobile. The goal, however, is the most accurate and mobile analyzers.
The simplest may not meet the expectations. Currently, there are attempts to select such a device (or a group of features that it must meet) that, in terms of the simplicity of its operation and the reliability of the presented results, can be used commercially at vehicle inspection stations. Devices that can potentially be included in this group can be categorized according to their principle of operation: (1) Aerosol electrometer particle counter (AEPC)-the measurement is performed using an electrometer. The charged particles cause the electrode to generate a current proportional to their quantity. (2) Condensation particle counter (CPC)-optical measurement of particles enlarged by condensation in a saturated aerosol (e.g., isopropanol). (3) Diffusion charger (DC)-electrically charged aerosol particles are analyzed, which flow into the condenser in a laminar form with air. There, the stream of charged particles is separated due to the electric voltage and fed to the electrode.
Devices for measuring particulate matter, used in the DPF system efficiency test, to be able to ensure repeatability and reliability of results and be widely used in various operating conditions, should meet a number of requirements: • provide the measurement result within the permissible limit errors or switch to the mode in which the measurement is impossible, which should be signaled by an appropriate message, • work in the full measuring range in the set and the widest possible range of temperatures, atmospheric pressure and humidity, • work and also show correct results in dynamic conditions (vibrations, shocks), as well as in the presence of an electromagnetic field, • meet the maximum permissible errors in the measuring range of 5000-500,000 cm −3 , and in the case of a variable, automatically switch over the measuring range and also meet the requirements when changing the range, • have adequate effectiveness in detecting particles with different diameters, • have a self-checking system for faults and errors, • have the possibility of zero and calibration control, sample volume control, etc., • allow for a simple evaluation of the results, • ensure a constant volume of the exhaust sample taken, • be equipped with a system for emptying the exhaust gas collection system, • be characterized by a short warm-up time and signal that the readiness time is reached, as well as an adequately fast response time.
Additionally, such devices should be secured against possible tampering (mainly by means of software protecting against changes or attempts to connect additional interfaces).
As can be seen, the problem of selecting a device that can be used in testing DPFs is complex and the devices, in order to be able to fully perform their task, should overcome many restrictions related to the method and accuracy of measurement, mobility, speed of operation and resistance to manipulation [16].
Therefore, work on defining the minimum requirements for devices for measuring particle number in such a way that they can be used on a mass scale is still underway.
This measurement is to give an unambiguous answer to the question of whether the diesel particulate filter is operational and whether it has not been removed. The value of emissions alone is not as important for periodic technical inspections as confirming the efficiency of exhaust gas treatment systems.
The authors of the study focused on the issue of measuring the number of particles with devices that could potentially be widely used in the testing of DPF systems.
Many solutions allowing the measurement of the number of solid particles ranging in size from several nanometers (in the approval tests, the lower limit of the tested particles is 23 nm) to several dozen micrometers were analyzed.
This limit corresponds to the size range of particles emitted by diesel engines [17]. The size distribution of solid particles emitted by diesel engines is shown in Figure 1. Two devices were selected for testing. Their cost will be very important for mass purchases by vehicle inspection stations. One device used a DC and the other a CPC. A series of comparative tests was carried out on a group of 50 vehicles aimed at verifying the hypothesis that the technical solutions used in selected devices will allow for their use in a reliable diagnosis of DPFs, carried out on a large scale during periodic technical inspection, which would solve the common problem of excessive emissions of vehicles with defective filters [18].

Methodology
A relatively inexpensive, less accurate method is the diffusion charger method. Measurement by this method can be divided into two stages: charging the aerosol particles with a single-pole DC and then measuring the current induced by the charged particles with a Faraday cage electrometer.
In the first stage, the corona discharge generates positive ions that diffuse to the particle surfaces, which gives the particles a charge directly proportional to their size. In the second stage, a current is measured that is proportional to the flow rate, the particle number concentration and the average charge per particle-therefore it depends on the particle size [19].
For diffusion chamber and induced current measurements, the dependence on particle size can be reduced by using a pulsed electro-filter placed in front of the Faraday cage during steady-state charging.
The relation of the size of the particles flowing through the filter is inversely proportional to the charging process, which allows the measurement of the number of particles with a size of 20-200 nm corresponding to the particles in the exhaust gas [13].
The current signal is directly proportional to the product of the concentration of the particulates and their diameter, so N can be determined if d is known. If the device is calibrated at d = 70 nm, then for typical diameters of 50-100 nm the device will have a maximum uncertainty of about 40% (according to the device manufacturer), which in fact is not much worse than the full PMP approval test, for which the uncertainty is about 25% [20]. An example of a diagram of a diffusion charger is shown in Figure 2. During the research, the Naneos Partector prototype was used. A unipolar charger is a charger that gives periodic pulses to alternately charge the suspended particles. These charges pass through a Faraday cage connected to an electrometer which gives indications by constantly keeping the entire cage electrically neutral. Thus, the load on the Faraday cage is always opposite to that inside the cage. By measuring the charge flowing into the cage, the charge per aerosol can be calculated. The electrometer signal is sinusoidal and its amplitude is a measure of the total charge on the particles and is calibrated for lung deposited surface area (LDSA) size distributions of particulate concentrations. Thanks to this approach, drifts of the electrometer zero offset are automatically compensated and changes in temperature or humidity have almost no effect on the measurement result. The device start-up time is very short, compared to others, and amounts up to 16 s [22].
This technique enables the non-contact measurement of the aerosol charge. The measurement is insensitive to changes in the probe position due to measuring only the signal amplitude. The above measuring method was used in the construction of the miniature device. The Naneos measures the charge transfer to the aerosol in the unipolar charging mode [23].
The improved version of the partition device is additionally equipped with an electrostatic precipitator that allows the collection of particles on the transmission electron microscopy (TEM) mesh for later analysis [23].
Another method is the one that involves a condensation particle counter. For the determination of particle concentrations in diesel engines, the most commonly used device during type-approval tests is the CPC. Upon entering the measuring chamber, the fumes as a stream of aerosol become saturated with alcohol vapors (Sensors Inc.'s Feldheider Str. 60, 40699 Erkrath, Germany APA uses pure isopropyl alcohol).
The mixture is cooled in the condenser tube. During this time, the vapor becomes supersaturated and condenses on the molecules. The effect of this phenomenon is that the particles increase their diameter to about 10 µm, which enables their optical detection [6,24]. The detection limit of the particles depends on the degree of saturation, which is extremely important when detecting the smallest particles (below 3 nm). The detection limit of modern CPCs is around 10 nm. Figure 3 shows one of the most technically advanced devices available on the NPET market according to TSI [25]. In this study, the CPC APA device by Sensors was used, the diagram of which is shown in Figure 4.

Results and Discussion
Two devices were used for measurements: Partector by Naneos, operating using the diffusion charger method, and Automotive Particle Analyzer (APA) by Sensors, using the condensation particle counter method. Used devices are shown on Figure 5.    The Table 1 presents tested vehicles obtained results. From the results obtained, it can be concluded that 10 vehicles may not be equipped with a DPF system, as it has been removed or is completely out of order. Twenty-three of the vehicles (almost half of the vehicles tested) had an inoperative DPF system. This illustrates the scale of the problem. The percentage of vehicles with defective DPF systems is shown in Figure 9. The evaluation criteria were negotiated and accepted for use by the NPTI working group [7]. Currently, Germany, the Netherlands, Belgium and Switzerland are at the stage of implementing particle number (PN) measurements into PTI. A value of 250,000 #/cm 3 is used as the diesel particulate filter evaluation criterion for vehicles equipped with Euro 5 and Euro 6 engines, while a value of 1,000,000 #/cm 3 is used for vehicles equipped with a diesel particulate filter with a Euro 4 engine.
Vehicles equipped with a compression ignition engine that do not have a diesel particulate filter typically emit more than 10,000,000 #/cm 3 , although there are cases where the measurement result is in the range of 2,000,000-10,000,000 #/cm 3 [7].
In our research, we obtained results as follows. For vehicles that have a defective or partially defective filter, their emission is usually in the range of 250,000-4,000,000 #/cm 3 .
The emission of particulate matter in vehicles with an efficient diesel particulate filter is several hundred #/cm 3 , although it can also be at the level of thousand #/cm 3 . Emissions in the range of tens of thousands indicates the initial problems with the diesel particulate filter.
The emission level is essential, not the specific measurement value. Measured values change over time and fluctuate. Under natural conditions, the number of particles in the environment (in the air) varies from about 1000 #/cm 3 up to 20.000 #/cm 3 .
The advantage of the measurement is that it takes place under steady-state conditions at an idle speed. The measurement result depends on the engine operating temperature, however, this dependence is weak enough to influence the measurement value, but the emission level does not change significantly. Similarly, the particulate matter emission depends on the engine rotational speed, and as the engine rotational speed increases the amount of particulate matter increases, but this increase does not significantly affect the emission level.
Two devices were used for the test, taking measurements using the two different methods described above. In all cases, the results of the measurements with both devices were at a similar level, and there was not a single case where one device gave a negative result and the other a positive result. Either a negative or a positive result was obtained.
It should be noted that the Swiss device Partector from Naneos (Alte Spinnerei 9, CH-5210 Windisch) is a prototype device and, as the manufacturer admits, imperfect. However, this prototype allows for unambiguous assessment the particulate filter efficiency. The advantages of the Naneos device are its mobility and simplicity as well as short measurement time, while its disadvantages are that it is limited to the measurements of compression-ignition engines and shows quite frequent signaling of measurement errors.

Conclusions
The results obtained indicate certain divergences. However, it should be remembered that the purpose is to define a "functional-not functional" DPF. Therefore, with this consideration, a detailed comparative analysis of the obtained results was not performed.
The APA by Sensors device is more accurate, has a more extensive measurement procedure and during the measurement, the engine speed and the temperature of the engine coolant are recorded. The disadvantages include the sensitivity of the device to transport and a complex structure, which significantly increases the risk of malfunctions.
Based on the results of measurements carried out with both devices, it is possible to answer a question that until now has not been possible to answer at a vehicle inspection station: whether the diesel particulate filter has been removed from the vehicle and, above all, whether it is operational. The way to perform the measurement is extremely simple and "friendly" compared to the free acceleration method as is used today.
The results of this study clearly recommend both measurement methods to be used at vehicle inspection stations.
Introduction of this new measurement method would allow effective elimination of the phenomenon of removing particulate filters from cars equipped with compressionignition engines and, at the same time, contribute to the improvement of air quality in large cities.