- Email: [email protected]
Tougher valve seats use two hard particles but no cobalt
technical trends
Tougher valve seats use two hard particles but no cobalt Japanese researchers have produced a new material for internal combustion engine valve seats that is more environmentally friendly, and cheaper... xhaust valve seat inserts (VSI) are often used in severe hightemperature, high-pressure and non-lubricant conditions. That is why cobalt-containing sintered materials (current material (A)), which have superior strength at high temperature and good solid lubrication, are mainly used for exhaust VSI [1]. However, environmental laws increasingly restrict the use of cobalt as well as lead. Cobalt is also expensive and has supply chain stability concerns. Addressing these factors, Nissan Motor Co and Hitachi Powdered Metals have developed and used a high-speed steel based sintered material for exhaust VSI (current material (B)) which is both cobalt and lead free [2]. Competition in the global market place imposed a final condition - a need to improve machinability to reduce the cost of the inserts. So in summary there was a need for the development of a new exhaust VSI that answered the following four needs; sufficient wear resistance in conventional engines; cobalt and lead free; good machinability characteristics and reduced production costs. There were some questions already raised by normal quality assurance testing. Plastic flow was partially observed on the wear surface of a valve seat during a durability test on a current exhaust VSI, and adhesion of part of the VSI material on the valve face was also seen. It was concluded that one of the factors of VSI wear was the adhesion caused
E
20
MPR April 2005
by sliding between the VSI and the valve face. It was also concluded that this adhesive wear was affected by the combustion temperature, the combustion pressure and/or the wear surface conditions such as the surface temperature of the VSI, lubrication or the state of the oxide film. So, in order to obtain sufficient adhesive resistance - wear resistance - it was necessary to select a matrix and hard particles with superior strength and hardness at high temperatures, good lubrication and the capability to form an oxide film. To investigate further, water-atomised powders (Table 1) and graphite were used as raw materials. Zinc stearate (0.5 mass %) was added as a lubricant. All materials were mixed for 1800 seconds by a Vtype mixer. Each mixture was compacted at a pressure of 640MPa into a predetermined shape. Sintering was carried out in a dissociated ammonia gas at 1433K for 3600 seconds. Wear resistance was evaluated by using a VSI wear tester. A valve that had been used in an actual engine and specimens machined to the shape of an actual VSI were used for the VSI wear tester.
Careful consideration was given to the high temperature hardness of the factors influencing the adhesive resistance. It was therefore decided to select a matrix with a high- temperature hardness which was higher than the current material (A). Through an advanced investigation, an Fe-Mo alloy was chosen as the matrix alloy, because the hardness of the Fe-Mo alloy was increased by the precipitated molybdenum carbide and there was only a small decrease in hardness at a high temperature. A suitable molybdenum level was also investigated. To determine the amount of molybdenum necessary in the alloy powder for a matrix, the high-temperature hardness and radial crushing strength was measured using an Fe-Mo alloy with molybdenum amounts varying from 1.5 per cent to 5.0 per cent. Specimens were prepared to the mixing ratios shown in Table 2. Figure 1 shows the relationship between the high-temperature hardness and the amount of molybdenum. The matrix hardness increased at both room temperature and at high temperature as the amount of molybdenum
Table 1: Chemical composition of raw material powder (mass%) Fe Cr Mo V Si C Fe-Mo Bal. 1.5 - 5.0 Fe-Cr-C
Bal.
12.0
1.0
0.8
0.4
1.5
Fe-Mo-Si
Bal.
-
35
-
1.4
-
Base iron powder Hard Hard particle particle
0026-0657/05 ©2005 Elsevier Ltd. All rights reserved.
Figure 1, The relationship between hightemperature hardness and the amount of m0lybdenum.
was increased. The hardness of Fe-Co, which is the matrix for a current material, (A) is also shown in Figure 1. The hardness of the Fe-Mo alloy with the molybdenum amount of 2.1 per cent or more was equal or higher to that of the Fe-Co used in the current material (A). Figure 2 shows the relationship between the radial crushing strength and the amount of molybdenum. The radial crushing strength increased with the increase in molybdenum in the range of 1.5 per cent to 3.5 per cent. However, a sample that included 5 per cent molybdenum had a decrease in radial crushing strength compared with that of a 1.5 per cent to 3.5 per cent molybdenum sample. It was assumed that the decrease in radial crushing strength was due to the precipitation of molybdenum carbides. Judging from these results, it was determined that the amount of molybdenum to be included in the alloy powder for the matrix was 3.5 per cent. Two types needed It was decided that two kinds of hard particles with different hardness (henceforth referred to as the first hard particle and second hard particle) should be dispersed in the investigated VSI material. With two types of hard particles, both hardness and strength should be achieved in the new VSI. Powders containing cobalt or ferroalloy powders such as Fe-Mo have generally been used as additional hard particles in past exhaust VSIs. A Co-Mo-Cr-Si alloy powder has also been used for the current material (A). This powder has cobalt-based hard particles forming hard phases consisting of mainly molybdenum silicide in the VSI material, and the hard phases have greatly contributed to the
metal-powder.net
Figure 2. Radian crishing strength increases with molybdenum addition - up to a point. Then it reduces due to carbide formation.
Figure 3. Adding more fe-Cr-C up to 25 per cent had little effect on radial crushing strength.
improvement in wear resistance. Based on the above, it was decided to select a hard particle containing molybdenum silicide but without cobalt. A new hard particle Fe-Mo-Si was therefore developed as a first hard particle (FHP) satisfying these requirements. The hardness of the Fe-MoSi was equal to that of a Co-Mo-Cr-Si alloy powder used for the current material (A).
10.0mass %) as the hard particle. Considering the severe conditions in which the VSI is used, it is necessary for the thermal conductivity to be as high as possible and the coefficient of the thermal expansion as low as possible. It was confirmed that the Fe-35mass%Mo1.4mass%Si was superior to Fe-35 mass%Mo-10mass%Si in both these properties. As a result of the results, the Fe-35Mo-1.4Si powder was selected as the first hard particle without cobalt.
Influence of Si The chemical composition of Fe-Mo-Si was determined based on the results of an investigation of the radial crushing strength, thermal conductivity, thermal expansion, proof strength, adhesion with the matrix and hardness of the hard particle phases. Especially, as it became clear in investigations that the amount of Si in the Fe-Mo-Si alloy greatly influenced several properties of the sintered body, the evaluation to decide a suitable Si amount was carried out. The amount of Mo in the particle was determined to be 35 mass% from the result of the advance research. Table 3 shows the thermal conductivity and the coefficient of thermal expansion of VSI materials using Fe-35Mo-xSi (x=1.4mass% or
Second hard particles There were two objectives to adding a second hard particle; one was to increase the amount of hard particles for the improvement of wear resistance, and the other was to maintain the strength of the new VSI. An advanced investigation was performed to choose a second hard particle which would satisfy these two requirements, and Fe-Cr-C was selected. Fe-Cr-C functioned as the hard phase in the new VSI, since the chromium carbide precipitated. And it was considered that the radial crushing strength increased because of the diffusion of the chromium in Fe-Cr-C. In addition, the diffusion of the chromium
Table 2: Mixing rate of test samples. (mass%) Base powder Hard particles Fe-Mo Fe-Cr-C Fe-Mo-Si
Gr.
Lub .
75.0
0.65
0.5
15.0
10.0
Table 3: Thermal conductivity and coefficient of thermal expansion Thermal conductivity Coefficient of thermal (Wm-1K-1) expansion (10-6) 1.4mass%Si 20 13.7 10.0mass%Si
16
14.5
April 2005 MPR
21
The authors
Figure 4. The relationship between wear and increased Fe-Mo-Si can be seen.
Figure 5. Radial crushing strength measured in relation to the amount of hard particles.
Table 4: Properties of new material Radial crushing strength at 573 K (MPa) New material >830 Current material
>780
Hardness HRA 60
Density (Mg/m3) 7.0
60
7.2
had a secondary effect other than the two objectives, and that was of increasing the matrix hardness facilitating the generation of an oxide film. Figure 3 shows the relationship between the radial crushing strength and the additional amount of Fe-Cr-C. There was little decrease in the strength between the additional amounts of Fe-Cr-C from 5 per cent to 25 per cent. It was possible to increase the amount of the total hard particles added to the new VSI while maintaining sufficient strength. The amount of additional hard particles was decided based on the results of an investigation of the radial crushing strength and wear resistance. Figures 4 and 5 show the results of the evaluation of wear resistance and radial crushing strength for various additional amounts of hard particles. It was seen that there was a tendency for the wear of the VSI to decrease, as the amount of Fe-Mo-Si was increased. The radial crushing strength also decreased
linearly, as the amount of Fe-Mo-Si was increased. It was seen that there was a tendency for the wear of the valve seat insert to decrease, as the amount of FeCr-C was increased between the ranges of 5 per cent to 15 per cent. However, there was little improvement at the point of radial crushing strength where there was an addition of 25 per cent. Judging from the results mentioned above, it was concluded that the most suitable addition of hard particles was 10 per cent of Fe-Mo-Si and 15 per cent of Fe-Cr-C. By means of selecting these two types of hard particles, it was possible to increase the total amount of hard particles by 25 per cent while maintaining a sufficient strength level. As a result, sufficient wear resistance was obtained. Photo 1 shows the microstructure of the new VSI material. The matrix is bainite, and the Fe- Mo-Si and Fe-Cr-C hard phases were dispersed throughout the matrix. Table 4 shows the mechanical
Figure 6. Microstructure of developed material.
Figure 7. The new material performed well in engine tests.
22
MPR April 2005
This feature is based on The Development of a Cobalt-Free Exhaust Valve Seat Insert, by Akira Fujiki*, Mitsushi Oyanagi*, Tomonori Miyazawa*, Hiroki Fujitsuka**, and Hideaki Kawata**, a paper given at the PM 2004 World Congress in Vienna under the auspices of the European Powder Metallurgy Association. *Nissan Motor Co Ltd, Powertrain Technology and Prototype Development Department **Hitachi Powdered Metals Co Ltd
References 1) K Suzuki, Y Ikenoue, H Endoh and M Uchino: New Sintered Valve Seats for Internal Combustion LPG Engines, Modern Development in Powder Metallurgy, Vol.18-21 (1988), pp.157170. 2) H Kawata, K Hayashi, K Ishii, K Maki, A Ehira and M Toriumi: The Development of a High Speed Steel Based Sintered Material for High Performance Exhaust Valve Seat Inserts, SAE Technical Paper 980328. properties of the new VSI material. The manufacturing process of the new VSI material was 1P1S, while the manufacturing process of the current material (A) was 2P2S. The density of the new material was lower in comparison with that of the current material (A); nevertheless the new material has a higher radial crushing strength and hardness compared to that of the current material (A). Machinability tests of the newly developed material were carried out using a CNC lathe under conditions identical to mass production. It performed better than current materials. And it performed well in durability tests in an engine (Figure 7). The wear resistance shown by the newe VSI material was equal to that of current material (A) and showed a steady wear pattern without loss of the hard particles or cracking of the matrix. Some plastic flow was observed.
metal-powder.net
Tougher valve seats use two hard particles but no cobalt Japanese researchers have produced a new material for internal combustion engine valve seats that is more environmentally friendly, and cheaper... xhaust valve seat inserts (VSI) are often used in severe hightemperature, high-pressure and non-lubricant conditions. That is why cobalt-containing sintered materials (current material (A)), which have superior strength at high temperature and good solid lubrication, are mainly used for exhaust VSI [1]. However, environmental laws increasingly restrict the use of cobalt as well as lead. Cobalt is also expensive and has supply chain stability concerns. Addressing these factors, Nissan Motor Co and Hitachi Powdered Metals have developed and used a high-speed steel based sintered material for exhaust VSI (current material (B)) which is both cobalt and lead free [2]. Competition in the global market place imposed a final condition - a need to improve machinability to reduce the cost of the inserts. So in summary there was a need for the development of a new exhaust VSI that answered the following four needs; sufficient wear resistance in conventional engines; cobalt and lead free; good machinability characteristics and reduced production costs. There were some questions already raised by normal quality assurance testing. Plastic flow was partially observed on the wear surface of a valve seat during a durability test on a current exhaust VSI, and adhesion of part of the VSI material on the valve face was also seen. It was concluded that one of the factors of VSI wear was the adhesion caused
E
20
MPR April 2005
by sliding between the VSI and the valve face. It was also concluded that this adhesive wear was affected by the combustion temperature, the combustion pressure and/or the wear surface conditions such as the surface temperature of the VSI, lubrication or the state of the oxide film. So, in order to obtain sufficient adhesive resistance - wear resistance - it was necessary to select a matrix and hard particles with superior strength and hardness at high temperatures, good lubrication and the capability to form an oxide film. To investigate further, water-atomised powders (Table 1) and graphite were used as raw materials. Zinc stearate (0.5 mass %) was added as a lubricant. All materials were mixed for 1800 seconds by a Vtype mixer. Each mixture was compacted at a pressure of 640MPa into a predetermined shape. Sintering was carried out in a dissociated ammonia gas at 1433K for 3600 seconds. Wear resistance was evaluated by using a VSI wear tester. A valve that had been used in an actual engine and specimens machined to the shape of an actual VSI were used for the VSI wear tester.
Careful consideration was given to the high temperature hardness of the factors influencing the adhesive resistance. It was therefore decided to select a matrix with a high- temperature hardness which was higher than the current material (A). Through an advanced investigation, an Fe-Mo alloy was chosen as the matrix alloy, because the hardness of the Fe-Mo alloy was increased by the precipitated molybdenum carbide and there was only a small decrease in hardness at a high temperature. A suitable molybdenum level was also investigated. To determine the amount of molybdenum necessary in the alloy powder for a matrix, the high-temperature hardness and radial crushing strength was measured using an Fe-Mo alloy with molybdenum amounts varying from 1.5 per cent to 5.0 per cent. Specimens were prepared to the mixing ratios shown in Table 2. Figure 1 shows the relationship between the high-temperature hardness and the amount of molybdenum. The matrix hardness increased at both room temperature and at high temperature as the amount of molybdenum
Table 1: Chemical composition of raw material powder (mass%) Fe Cr Mo V Si C Fe-Mo Bal. 1.5 - 5.0 Fe-Cr-C
Bal.
12.0
1.0
0.8
0.4
1.5
Fe-Mo-Si
Bal.
-
35
-
1.4
-
Base iron powder Hard Hard particle particle
0026-0657/05 ©2005 Elsevier Ltd. All rights reserved.
Figure 1, The relationship between hightemperature hardness and the amount of m0lybdenum.
was increased. The hardness of Fe-Co, which is the matrix for a current material, (A) is also shown in Figure 1. The hardness of the Fe-Mo alloy with the molybdenum amount of 2.1 per cent or more was equal or higher to that of the Fe-Co used in the current material (A). Figure 2 shows the relationship between the radial crushing strength and the amount of molybdenum. The radial crushing strength increased with the increase in molybdenum in the range of 1.5 per cent to 3.5 per cent. However, a sample that included 5 per cent molybdenum had a decrease in radial crushing strength compared with that of a 1.5 per cent to 3.5 per cent molybdenum sample. It was assumed that the decrease in radial crushing strength was due to the precipitation of molybdenum carbides. Judging from these results, it was determined that the amount of molybdenum to be included in the alloy powder for the matrix was 3.5 per cent. Two types needed It was decided that two kinds of hard particles with different hardness (henceforth referred to as the first hard particle and second hard particle) should be dispersed in the investigated VSI material. With two types of hard particles, both hardness and strength should be achieved in the new VSI. Powders containing cobalt or ferroalloy powders such as Fe-Mo have generally been used as additional hard particles in past exhaust VSIs. A Co-Mo-Cr-Si alloy powder has also been used for the current material (A). This powder has cobalt-based hard particles forming hard phases consisting of mainly molybdenum silicide in the VSI material, and the hard phases have greatly contributed to the
metal-powder.net
Figure 2. Radian crishing strength increases with molybdenum addition - up to a point. Then it reduces due to carbide formation.
Figure 3. Adding more fe-Cr-C up to 25 per cent had little effect on radial crushing strength.
improvement in wear resistance. Based on the above, it was decided to select a hard particle containing molybdenum silicide but without cobalt. A new hard particle Fe-Mo-Si was therefore developed as a first hard particle (FHP) satisfying these requirements. The hardness of the Fe-MoSi was equal to that of a Co-Mo-Cr-Si alloy powder used for the current material (A).
10.0mass %) as the hard particle. Considering the severe conditions in which the VSI is used, it is necessary for the thermal conductivity to be as high as possible and the coefficient of the thermal expansion as low as possible. It was confirmed that the Fe-35mass%Mo1.4mass%Si was superior to Fe-35 mass%Mo-10mass%Si in both these properties. As a result of the results, the Fe-35Mo-1.4Si powder was selected as the first hard particle without cobalt.
Influence of Si The chemical composition of Fe-Mo-Si was determined based on the results of an investigation of the radial crushing strength, thermal conductivity, thermal expansion, proof strength, adhesion with the matrix and hardness of the hard particle phases. Especially, as it became clear in investigations that the amount of Si in the Fe-Mo-Si alloy greatly influenced several properties of the sintered body, the evaluation to decide a suitable Si amount was carried out. The amount of Mo in the particle was determined to be 35 mass% from the result of the advance research. Table 3 shows the thermal conductivity and the coefficient of thermal expansion of VSI materials using Fe-35Mo-xSi (x=1.4mass% or
Second hard particles There were two objectives to adding a second hard particle; one was to increase the amount of hard particles for the improvement of wear resistance, and the other was to maintain the strength of the new VSI. An advanced investigation was performed to choose a second hard particle which would satisfy these two requirements, and Fe-Cr-C was selected. Fe-Cr-C functioned as the hard phase in the new VSI, since the chromium carbide precipitated. And it was considered that the radial crushing strength increased because of the diffusion of the chromium in Fe-Cr-C. In addition, the diffusion of the chromium
Table 2: Mixing rate of test samples. (mass%) Base powder Hard particles Fe-Mo Fe-Cr-C Fe-Mo-Si
Gr.
Lub .
75.0
0.65
0.5
15.0
10.0
Table 3: Thermal conductivity and coefficient of thermal expansion Thermal conductivity Coefficient of thermal (Wm-1K-1) expansion (10-6) 1.4mass%Si 20 13.7 10.0mass%Si
16
14.5
April 2005 MPR
21
The authors
Figure 4. The relationship between wear and increased Fe-Mo-Si can be seen.
Figure 5. Radial crushing strength measured in relation to the amount of hard particles.
Table 4: Properties of new material Radial crushing strength at 573 K (MPa) New material >830 Current material
>780
Hardness HRA 60
Density (Mg/m3) 7.0
60
7.2
had a secondary effect other than the two objectives, and that was of increasing the matrix hardness facilitating the generation of an oxide film. Figure 3 shows the relationship between the radial crushing strength and the additional amount of Fe-Cr-C. There was little decrease in the strength between the additional amounts of Fe-Cr-C from 5 per cent to 25 per cent. It was possible to increase the amount of the total hard particles added to the new VSI while maintaining sufficient strength. The amount of additional hard particles was decided based on the results of an investigation of the radial crushing strength and wear resistance. Figures 4 and 5 show the results of the evaluation of wear resistance and radial crushing strength for various additional amounts of hard particles. It was seen that there was a tendency for the wear of the VSI to decrease, as the amount of Fe-Mo-Si was increased. The radial crushing strength also decreased
linearly, as the amount of Fe-Mo-Si was increased. It was seen that there was a tendency for the wear of the valve seat insert to decrease, as the amount of FeCr-C was increased between the ranges of 5 per cent to 15 per cent. However, there was little improvement at the point of radial crushing strength where there was an addition of 25 per cent. Judging from the results mentioned above, it was concluded that the most suitable addition of hard particles was 10 per cent of Fe-Mo-Si and 15 per cent of Fe-Cr-C. By means of selecting these two types of hard particles, it was possible to increase the total amount of hard particles by 25 per cent while maintaining a sufficient strength level. As a result, sufficient wear resistance was obtained. Photo 1 shows the microstructure of the new VSI material. The matrix is bainite, and the Fe- Mo-Si and Fe-Cr-C hard phases were dispersed throughout the matrix. Table 4 shows the mechanical
Figure 6. Microstructure of developed material.
Figure 7. The new material performed well in engine tests.
22
MPR April 2005
This feature is based on The Development of a Cobalt-Free Exhaust Valve Seat Insert, by Akira Fujiki*, Mitsushi Oyanagi*, Tomonori Miyazawa*, Hiroki Fujitsuka**, and Hideaki Kawata**, a paper given at the PM 2004 World Congress in Vienna under the auspices of the European Powder Metallurgy Association. *Nissan Motor Co Ltd, Powertrain Technology and Prototype Development Department **Hitachi Powdered Metals Co Ltd
References 1) K Suzuki, Y Ikenoue, H Endoh and M Uchino: New Sintered Valve Seats for Internal Combustion LPG Engines, Modern Development in Powder Metallurgy, Vol.18-21 (1988), pp.157170. 2) H Kawata, K Hayashi, K Ishii, K Maki, A Ehira and M Toriumi: The Development of a High Speed Steel Based Sintered Material for High Performance Exhaust Valve Seat Inserts, SAE Technical Paper 980328. properties of the new VSI material. The manufacturing process of the new VSI material was 1P1S, while the manufacturing process of the current material (A) was 2P2S. The density of the new material was lower in comparison with that of the current material (A); nevertheless the new material has a higher radial crushing strength and hardness compared to that of the current material (A). Machinability tests of the newly developed material were carried out using a CNC lathe under conditions identical to mass production. It performed better than current materials. And it performed well in durability tests in an engine (Figure 7). The wear resistance shown by the newe VSI material was equal to that of current material (A) and showed a steady wear pattern without loss of the hard particles or cracking of the matrix. Some plastic flow was observed.
metal-powder.net