Dr.-Ing. M. Bambach / Prof. Dr.-Ing. G. Hirt (institute of metal forming (IBF), RWTH Aachen)
The nowadays development of high strength steel alloys for light-weight construction resulted into qualities with high manganese content with aluminium and silicon doping. Within these alloys the mechanical properties are mainly influenced by their chemical composition. These alloying elements adjust the stacking fault energy that determines the activated forming mechanism like TRIP (TRansformation Induced Plasticity) and TWIP (TWinning Induced Plasticity) and leads to steels with high strength values and good formability. Therefore the development uses more and more the metal physical phenomenons, but still trial and error tests to advance materials' properties dominate the today's development of alloys. In this collaborate research project model-based simulation tools for alloy development shall be generated using the Fe-Mn-C ternary system. Combining operator's needs with the ab-initio methods application relevant properties shall be referred to metal physical phenomenons and optimised model-based in the future.
The tasks of sub-project B2 are on the one hand, to provide the other sub-projects with test material and on the other hand the process development for hot forming.
Concerning the material distribution, B2 is the subsequent process step after B1 which provides the casted blocks. In B2 these blocks are shaped (e.g. by forging, hot rolling, cold rolling) and can be distributed on demand. During these hot forming operations the material is homogenized and the micro segregations of the manganese content in the cast state are reduced. Before the first step, forging, inhomogeneities in the top and bottom areas of the casted block are cut off. The following process step is prepared by welding on a tag which can be grabbed by the forging robot in order to reduce losses in test material. The block, which is about 50 cm long (14*14*50 cm3), gets heated up to 1150°C in the furnace and gets taken out by the robot. By forging the size is reduced within three passes to 5*16*120 cm3. The surface changes its colour from the silver-gray of the casted block to black-gray of the forged piece because of scaling. Scaling (which means the formation of an oxidised layer) is especially distinct in high manganese steels. Partly this layer chips off during forging and cooling.
The block is prepared for further processing by cutting off the tag and dividing it into slabs suitable for rolling after which annealing takes place. For this, the slabs are reheated to 1150°C and annealed for five hours. After cooling in free atmosphere, homogenisation is completed. Before rolling the slabs are reheated in a furnace another time and held there for 10 minutes in order to heat trough (=even temperature in the whole volume). The slabs are now taken out by hand with callipers and put in the rolling mill. After every pass the work-piece is reheated again in the oven. The thickness is reduced from 50 to about 2 mm within 10-15 passes of hot rolling in a two-high rolling stand.
The desired thickness of the plate is achieved by the repetition of rolling, the so called pass-schedule. The number of passes is determined by the possible height-reduction per pass. The height-reduction is basically limited by two factors: the bite condition and the maximal rolling force. The bite condition is not fulfilled when the proportion of work piece to rolling-gap is too high. In this case the rolls can’t ‘bite’ the material because friction is too small. The maximal rolling force is limited by the mill stand. When there is too much thickness-reduction per pass, the maximum allowable rolling force can be exceeded.
The hot-rolled products are covered with an oxid-layer of partly rolled-in scale. By using a descaler (high pressure water jet) before each rolling pass the scale can be removed. This way the effect can be minimised. Scale-layers that form during the cooling between single passes can be removed before cold rolling by glas shot peening.
As the higher strength of the material at room-temperature requires higher stresses for cold rolling, the contact-area between the rolls and the work-piece is reduced using smaller rolls. With a smaller roll diameter the deflection of rolls increases because of the loss of stiffness. By using so called back-up rolls above and below the working rolls, which means rolls with a bigger diameter to support the working rolls (four-high mill), the roll reflection can be reduced.
The process development is divided in two areas, the numerical simulation and the physical simulation. Numerical simulation tries to recreate and analyse a forming-processes with highest accuracy by mathematical equations to increase the understanding of the process and of the resulting product-properties. For the simulation several input parameters are needed. B2 receives these from partner-projects (e.g. heat capacity, heat conduction, density) or ascertains them in self-conducted principle tests (e.g. flow stresses).
Such principle tests are carried out to determine the emission coefficient and the heat transfer as well. The major focus is on the flow stresses in dependence of temperature and strain-rate to describe the material behaviour during forming. At the Institute of Metal Forming the flow stresses are derived from upsetting tests using cylindrical samples. The cylindrical samples are equipped with lubrication slots on the faces to reduce the friction during the test and allow for homogenous strain distribution. The parameters varied are temperature and strain-rate. To reduce the loss of temperature of the small samples during analysis, the whole setup is heated to the respective forming temperature. The relevant parameters flow stress and strain are derived from the measured force-distance curves.
The physical simulation is carried out on small samples which show high similarity to the imaged real process in stress and strain distribution. The material-properties are affected by the chemical composition and the history of the material which means the course of temperature and time and the kind of deformation. The results from the physical simulation tests can be analogously transferred to industrial scale using similarity rules. Important here are - next to the influence of the parameters on the forces during deformation - the material-properties (e.g. grain size) adjusted by the deformation-parameters.
The challenges of the project lie mainly in the characterisation and process chain development of high manganese steels. To determine the material properties during forming processes the described upsetting tests, in dependence of temperature and strain rate, are conducted for several high manganese compositions.
The diagrams show material behaviour of two different Fe-Mn-C-alloys with 23% manganese and 0.3 respectively 0.6% carbon. The hot forming of these alloys at low temperatures (700°C) resembles the flow behaviour of usual austenitic chrome-nickel-steel X5CrNi 18-10 (AISI 304 / 1.4301). The influence of carbon is visible as the flow curve rises slightly with an augmentation of carbon content. At a temperature of 1200°C the measured flow curves of both Fe-Mn-C alloys are lower than the ones of the comparison-material and the influence of carbon content is not noticeable any more. The difference between the flow curves of the Fe-Mn-C alloys and the comparison-steel 1.4301 increases at higher temperatures.
As a result these Fe-Mn-C alloys can – regardless of their outstanding properties at room-temperature - be hot-formed with lower forces than Cr-Ni-steels in industrial hot-forming processes such as hot rolling. This effect is supported by a decrease of flow stress of these Fe-Mn-C alloys by a factor of 4-6 at a temperature-rise from 700 to 1200°C.