2016 Volume 63 Issue 7 Pages 543-547
We aimed to develop a new powder metallurgy process to produce next generation common rail diesel nozzles with tiny and complexly shaped holes. High-speed centrifugal compaction and powder metallurgy were used. Three nozzle samples with holes of straight and tapered (8°, 16°, enlarged inwardly) were made. Three samples without external and inner defects were examined by spray test, which was performed with an ambient pressure of 1 and 4 MPa, an injection pressure of 150 MPa, and was studied by filming using a high-speed camera at 8000 fps. The measuring parameters were the penetration, spray angle, and spray cone angle from the spray aperture. As the taper angle increased, the penetration and spray cone angle increased.
Diesel engines exhibit many advantages compared with gasoline engines, such as a high torque at low-speed and a low fuel consumption because of high thermal efficiencies, and therefore receive much attention for use in 21st-century engines. Diesel engines struggle to reduce exhaust pollutant matter, especially particulate materials and NOx gases, which restricts their use. A new solution must be established to reduce simultaneous emissions of such matters1,2).
In nozzle tip designs, smaller spray apertures of less than 100 μm are a primary requirement. A spray with high cone angle is also favorable to provide uniform and completely evaporated fuel throughout the cylinder at the time of ignition. However, a countereffect exists in decreasing the penetration of the spray when spray holes are straight. Therefore, a special nozzle design such as tapered holes is required to improve penetration with small spray holes.
We produced nozzle tips with multiple tiny tapered holes. The taper must be enlarged inwardly which is difficult to achieve using current drilling techniques. Therefore, we chose a newly developed powder metallurgy technique. The powder was compacted by a high-speed centrifugal compaction process (HCP), and multiple tapered holes were formed with resin cores that are eliminated during the process. Resin cores were made from ultraviolet-hardening resin using a three-dimensional printer3).
This report succeeds a former report4) that focuses on the properties of processed nozzles, shapes, microstructures, and spray properties.
A fine (2.56 μm) steel powder SCM415 (Epson Atmix Co., Japan) was used as starting material (Table 1). The powder was prepared as a non-aqueous slurry and was compacted by the HCP that buries 3D-printed resin core in the compact. It was heated in a vacuum to eliminate the resin mold and core by thermal decomposition and was sintered with further heating. The process detail is provided in a previous report4). In this process, three different nozzles are fabricated with i) straight, ii) low-tapered (8°), and iii) high-tapered (16°) holes. The properties of the nozzles produced by this process are evaluated in this study.
| Chemical composition [mass%] | Average particle size [μm] | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| C | Si | Mn | P | S | Ni | Cr | Mo | Cu | Fe | |
| 0.398 | 0.20 | 0.66 | 0.018 | 0.004 | 0.03 | 1.01 | 0.23 | 0.01 | Bal. | 2.56 |
Samples were cut vertically in half, and cross sections were studied using optical microscope (MODEL OPTIPHOTO-2, NIKON Co., Japan) and SEM: scanning electron microscope (S-3000H made by Hitachi, Ltd., Japan). Image analysis was used to determine the pore structure (size and amount) using software pickmap.
(b) Dimensional changeThe outer contour of the sintered nozzles was measured using a reading microscope (TOOLMAKER’S MICROSCOPE, Mitutoyo Co., Japan). The measured contour line was compared with the original shape design.
(c) Non-destructive inspectionFor samples to be used for spray tests, an X-ray transmission inspection was performed to avoid nozzle breakage at the time of the spray test. The inspection equipment are MTT-225, Shimadzu Co., japan, K-2, SOFTEX CO, Ltd., Japan, MSF-3M, Rigaku Co., Japan.
2.3 Spray testThe spray properties of the three nozzles were studied by high-speed camera (Phantom Miro, Research Vision Co.). Table 2 shows conditions of the fuel injection experiment. The frame rate of the high-speed camera was 8000 fps, that is, a still image was taken every 0.25 ms from spray start to end.
| Test fuel | Diesel Fuel (JIS No. 2) | |
|---|---|---|
| Ambient temperature Ta [K] | 300 | |
| Ambient pressure Pa [MPa] | 1.0 | 4.0 |
| Ambient density ρa [kg/m3] | 11.6 | 46.5 |
| Injection pressure Pinj [MPa] | 150 | |
| Injection period tinj [ms] | 3.2 | |
| Frame rate [fps] | 8000 | |
| Exposure time [μs] | 2 | |
Penetration, spray angle, and spray cone angle were measured from each still image (Fig. 1). Fig. 3 shows an elongated image of the spray aperture. In Fig. 3, “Z” is the average value of three spray apertures measured from SEM images and the actual diameter D is calculated using Eq. 1. Spray cone angle is the angle at a position of one hundred times D from spray aperture. Some data correction is conducted to obtain a true value for each data point, since the spray is inclined at 12° towards the bottom view, as shown in Fig. 2. Image analysis was conducted by software ImageJ and VirtualDub to contrast of the data, in which the base image (before spraying data) was subtracted from each still data at each time of spraying.
Definition of measurement points (X, Y).
Relationship between S and X, 100D and Y.
Relationship between Z and D.
| Eq. 1 |
SEM images of sintered compacts at various sintering temperatures (Ts) are shown in Fig. 4. Fine and homogeneous microstructures formed, although remaining pores of several microns existed, regardless of the Ts. The pores at lowest Ts (1273 K) were still elongated to the aspect ratio of two, and became spherical as Ts increased to 1408 K or more. Nonetheless, the porosity measured from these images is almost unchanged regardless of the Ts (Fig. 5). The pre-sintered sample density is about 63 % to the theoretical. In total, nozzles made by the HCP can be sintered at low temperature (1273 K) in solid-phase sintering. The use of fine starting powder may improve their sinterability.
Comparison of pore and sintering temperatures.
Relationship to density.
Contour lines of the sintered nozzles were measured (Fig. 6). They exhibited a large “linear” shrinkage of about 14 % on average, which is the nature of high-density sintering. The linear shrinkage of 14 % coincided with the “volumetric” density change from 63 % (in green compacts) to 98 % (in sintered compacts). Therefore, we can conclude that the entire product is sintered sufficiently throughout the product.
Shrinkage ratio of sintering body.
The provided images are the horizontally sliced cross sections as shown in Fig. 7. No cracks or other defects are found. All nozzle holes (each product contains three) were penetrated completely, although sometimes some residues (of possibly incompletely decomposed resin cores) existed in holes with minor contrast. In total, we conclude that the nozzles were fabricated successfully and so, we performed a spray test.
Results from X-ray test: a) straight nozzle, b) low-taper nozzle, c) high-taper nozzle.
Figs. 8 to 9 show typical examples of the appearance of the spray test for straight and high-tapered nozzles at an ambient pressure of 1 MPa, respectively. Three sprays in Fig. 8, as example, have different sizes, namely penetrations. The differences may be introduced by partial stuffing of core residue. We used the average data of three sprays for each measured value.
Fuel injection experiment with straight nozzle, Pa = 1.0 MPa.
Fuel injection experiment with high-taper nozzle, Pa = 1.0 MPa.
Spray diameters at the nozzle outlet were measured from the spray observations, and values of 0.21 mm, 0.16 mm, and 0.20 mm were obtained for the straight, low-tapered, and high-tapered nozzles, respectively. The nozzle diameter may affect the spray behavior, but we do not take this information into account in this research because the nozzle diameters are comparable.
(i) Spray-tip penetrationSpray-tip penetrations are summarized in Fig. 10. Although only partial data were taken because the spray passes over the frames at the midst of spraying, a general tendency exists for larger penetrations with tapered nozzles. When the ambient pressure increases from 1 to 4 MPa, the penetration becomes low, regardless of nozzle shape.
Spray tip penetration: a) Pa = 1 MPa, b) Pa = 4 MPa.
Spray angles were also taken partially since the data need penetration values to be defined (Fig. 11). In general, the early spray angle is higher, and it then decreases with time. The final spray angle at late timing appears to reach an ultimate constant value, and it increases with higher ambient pressure. No clear tendency exists, however, among the nozzle hole angles. Usually, the spray angle is easy to be effected by neighboring air. It is possible that differences in these results are masked by this effect.
Spray angle: a) Pa = 1 MPa, b) Pa = 4 MPa.
The spray cone angles exhibit contrary characteristics to spray angles, that is, a lower cone angle early when spraying, which increases with time (Fig. 12). A noteworthy characteristic of the spray cone angle is that the cone angles with tapered nozzle holes have increased over the straight one. Although the effect of nozzle hole angle (8 vs. 16°) is still ambiguous, certain advantages were noted for tapered nozzles with respect to spray cone angle. The effect of ambient pressure on the spray cone angle is the same as that of the spray angle, that is, final values increase with a higher ambient pressure. In contrast to the spray angle, the spray cone angle is not affected easily by neighboring air. Therefore, it may clarify the effect of nozzle hole shape.
Spray cone angle: a) Pa = 1 MPa, b) Pa = 4 MPa.
We aimed to develop a new powder-metallurgy process to produce next-generation common rail diesel nozzles with tiny, tapered, and multiple holes. The HCP was chosen and nozzle holes were formed using resin cores made with a 3D printer. The core was eliminated by thermal decomposition. Nozzles with three straight and tapered holes (8° and 16°) were formed. Spray tests with a common rail system were performed using these nozzles.
The results are as follows:
The authors acknowledge the valuable help for spray test by Dr. Yoshio Zama, division of mechanical science and technology, Gunma University.
