Improvement of die service life through local overlay welding

In this thesis, economical improvement of service life of a steel forging die to be used in closed die hot forging process was tried to be provided. In the light of information related to the failure history of the chosen forging die, overlay welding was locally applied to the regions on which most probable failure mechanism could take place. As compared to traditional methods in which whole gap of forging die is filled up with welding material, faster and more economical improvement in die service life was aimed. Before applying welding, samples were prepared by depositing each welding electrode on steel substrate having quality of DIN 1.2714. Two Co-based(Co1 and Co2) and two Fe-based(Fe1 and Fe2) hardfacing electrodes were used for sampling. All samples were prepared in accordance with welding procedure of hardfacing alloys.Maximum attention was paid during welding in order to keep process requirements. Because pre-heating of forging die, maintaining the neccesary minimum temperature level during welding, stress relieving and slow cooling after welding are very important for hardfacing. Six samples having dimensions of 30x30x6 mm from each type of welded substrate and pure substrate (1.2714) were produced. Welding beads were deposited in grooves having ~ 3,5 mm depth on substrate. Samples thicker than the others were subjected to machining in order to have uniform dimensions for each sample. By using the samples prepared, following inspection tests like hardness tests, microstructral characterizations, X-Ray diffraction analysis, wear tests at room temperature and high temperatures (400 °C ve 600 °C) were implemented in order to be able to analyse the performance of each welding deposit and also pure substrate material. The hardness value of Fe1 was measured as 49 HRC whereas Fe2 was 61 HRC. In order to be able to interpret that difference in terms of hardness values, microstructural characterizations were performed. According to images obtained under optical microscope, Fe2's structure consists of almost completely martensite accompanying with little retained austenite and α-Fe. Regarding to Fe1's microstructure, more α-Fe (white phases seen like a network) and retained austenite( white spots in martensite) were seen in comparison with Fe2. The existence of those phases were also verified by XRD characterization. The hardness values of Co1 and Co2 were measured as 39 HRC and 55 HRC, respectively. In order to be able to interpret that difference in terms of hardness values, microstructural characterizations were performed. Microstructural characterizations revealed that both Co1 and Co2 had α-Co matrix with eutectic carbides(〖Cr〗_7 C_3 , 〖Cr〗_23 C_6).The results of XRD characterizations also showed the same phases for both structures. After searching the reason of hardness difference despite similar XRD results, it has been noticed that the subsequent machining of Co2 surface caused such a difference. The reason related to difference of hardness was attributed to transformation of α-Co (FCC) to harder ε –Co(HCP) due to effect of subsequent additional maching while sampling. It is also a transformation process from FCC to HCP. Fe and Ni elements whose amounts are quantitatively more in Co1 than Co2 have also effect on lower hardness value due to their good FCC stabilizing property. Ball on disc wear tests for Fe1, Fe2 and substrate at three different temperatures (room temperature, 400°C and 600°C) under 4N load were performed. Test results showed that Fe1 had the best wear resistance while the substrate had the worst at room temperature. This behaviour was attributed to different strength values of oxide layers formed during test. Fe1 and substrate behaved similarly at 400 °C whereas Fe2 had much more wear loss in comparison with the rest. It is supposed that fragmentation of thin and less adherent oxide layer formed at the surface of Fe2 at 400°C caused such a big amount of material loss. Even though Fe2 has also similar chemical composition with Fe1, %4 Mo which does not exist in Fe2 encouraged the formation of thicker and stronger oxide layer on Fe1's surface. At 600 °C, an obvious improvement in terms of wear performance of Fe2 was observed. It is supposed that thick and hard oxide layer formed on the surface of samples, supported the resistance of layer against wear. The wear loss for Fe1 and substrate were more in comparison to 400 °C. It is supposed that the reason could be slightly softening of materials while approaching to annealing temperature with increasing temperature. Similarly, ball on disc wear tests for Co1 and Co2 at three different temperatures (room temperature, 400°C and 600°C) under 4N load were performed. Test results showed that Co2 had best wear resistance while the substrate had the worst at room temperature. This behaviour was attributed to different strength values of oxide layers formed during test. At 400 °C, Co2 had the worst wear resistance while the substrate materaial had the best among three. The difference between Co1 and Co2 was attributed to strenghts of oxide layers and amount of Co in their compositions. It is supposed that amount of formed thin and less adherent oxides was more in Co2 than that of Co1 due to excess amount of Co in the structure. At 600 °C, wear pefrormance of Co1 and Co2 apparently improved. It is attributed to formation of thick and hard 〖Cr〗_2 O_3 ve CoO oxide layers at temperatures above 400 °C and their lubricating effect against wear. Again due to excess amount of Co in Co2, the amount of thick and 〖Cr〗_2 O_3 ve CoO oxide layers was also quantitatively more than that of Co1, leading to a better wear performance than Co1 correspondingly. According to the extracted results from the inspection tests, the behaviour of each type of sample at different temperature and variations with temperature were determined. Since the temperature of forging die can reach up to 700 °C while in service, the results at 600°C were essentially taken into account due it's proxmity to real service temperature. The results at 600°C showed that Co2 sample had best wear resistance. Even it seemed obviously that Co2 must be used for hardfacing according to test results, it was not preferred due to it's high hardness value ( 55 HRC). Since the forging die is to be machined subsequently after welding process, high hardness values of die surface would cause serious difficulties during machining. The preparation period of die would take considerably longer time and the consumption of expensive machining tools (carbides) would be much more, consequently causing high costs. That's why Co1 whose hardness value is 39 HRC was preferred by considering the convenience for machining and it's characteristic feature. This characteristic feature of Co1 is abilitiy of being work hardened at high temperatures. By means of that , the subsequent machining can be performed easily thanks to low hardness value right after welding. Besides, the neccesity of a harder surface can be met at high temperatures. Event though, Fe1 electrode showed a similar wear resistance like Co2 at 600°C, it was even not assumed to use in application by considering the difference of real forging mechanism in respect to experimental wear mechanism. Since the reaction of surface against impact is very important for forging process, there must be not only hard surface, but also tough surface. When the failure mechanism history of the chosen forging die was analysed, it was seen that forging die failed to be in service due to the wear occured in flash line. This situation also shows itself on the finished part with a broadened flash line.It is also logical to assume that flow of hot steel through narrow flash line caused deficiencies on the surface of flash line. In reference to this observation, flash line of a selected forging die was machined 3,5 mm deeper than original depth of flash line during die preparation process; and this depth was filled up with Co1 welding electorde in accordance with fundemental six steps for overlay welding (pre-heating,welding,equilibrating,slow cooling,stress relieving and final slow cooling). After welding application, the forging die was subjected to additional machining process during ~ 1 hour in Cnc machine in order make it ready for production. After all above mentioned processes, it was seen that number of forged part till failure of forging die increased from 1250 pcs to 1850 pcs, leading 48 % better service life in comparison with that of raw 1.2714. Longer service life means that number of parts produced will be more and unit die cost will be less correspondigly. Besides , number of setups for the production will be less and effect of setup costs in unit cost will be less accordingly. Of course local overlay welding is a more economical solution than standard overlay welding of forging die in which whole die gap is filled up with welding material. But it is still an additional process for forging dies which causes emergence of additional costs. The bigger forging part is, the greater ratio of forging die cost in forging part's unit cost is. That's why the cost improvement thanks to local overlay welding application will be more for bigger parts. The key point is to be able to decide if local overlay welding process really makes sense for a part or not.