The application of CAE in superalloy fields at HTMRI CISRI
Feng Di Z.Long J.Hu J.Wu
High temperature materials Research Institute
Central Iron and Steel Research Institute. Beijing 100081. China
Abstract
The computer aided engineering (CAE) is impacting on traditional industry. Superalloys being used as high temperature structural materials of aeroengine are facing the competition of composites. In consideration of performance and cost benefits, the superalloy research and development are being sought through use of CAE to advance its affordability. In this paper, some efforts of using CAE to improve the cast and wrought superalloy affordability are introduced.
Introduction
Superalloy is one branch of traditional metallurgy industry. The evolution of the gas turbine engine depended on superalloy being developed in the past century. Although in the end of 20 century, composites and intermetallics for high temperature structural using have been developed rapidly, the key and main parts of jet engines will be still made of superalloys owing to the factor of cost and reliability. The demands of engine designer on the materials as key components are high performance, reliability and low cost. The affordability of superalloys is becoming increasingly important. In order to meet these challenge, computer aided engineering (CAE) is being sought to advance the performance of materials, reduce defects rates, manufacture cost and cycle time, and extend service life at HTMRI.
Computer Aided Engineering in Superalloy Investment Casting
Superalloy Investment Castings are high performers in today’s jet engine world and are the backbone of the aircraft engine industry.
Requirements for Superalloy castings in jet engine:
Light weight, high reliability, high quality
---- Thin section, integrated, and complicated structure
---- High dimension precision, low or no allowance
The shortcomings of conventional investment casting process
Lack of effective information exchange between designer and manufacturer.
Very difficulty to design and manufacture the tooling for complicated casting.
Long lead time, high cost
Especially high expense for design modifying
Over the last several years, the CAE in investment process research and development has been implemented at HTMRI to meet the requirements of the complicated superalloy castings in advanced jet engines. It has had resounding effects on key aspects of producing a complicated casting.
The flowchart of complicated casting development is follows and an example of CA Invest casting is presented in Fig.1:
(1) 3-D solid modeling (2) wax pattern (3) pouring assemble (4) casting
Fig 1 An example of CA investment casting
At first, the teamwork of “design-material-process” has been realized through using CAE in conventional investment cast process. The engine designer, materials researcher and castings engineer are working together, to discuss and solve the demands from designer, the performance of materials and the process of casting through electronic data. The 3-D computer aided design software are used. The transfer of the data into real components is being achieved through electronic data. A rapid prototyping pattern is used directly as investment pattern in shell making for complicated castings. It is very simple to change design pattern produced from electronic data, but much more difficult to modify hard tooling. The lead-time is shortened significantly, and the cost is reduced.
The traditional investment casting process is not suitable to the new demand of aeroengine complicated castings.
Computer aided engineering has had a significant impact on superalloy investment casting. The impact not only reduced lead time and cost, but also promoted a new teamwork mode and created an opportunity for engine designer to design more complicated casting that never be imagined.
The Hot Deformation Modeling of As-Cast Highly Strengthened Superalloys
The ingot conversion process is a key processing step in forging process especially for large diameter superalloy ingot and the highly strengthened superalloys such as GH742, Udimet 720 etc. which are widely used in new generation jet and land base power engines.
Highly strengthened superalloys have less hot workability caused by high level of alloying elements and ’ precipitating phase which induce the element segregation and metallurgy defects. Additionally, the hot workability is very sensitive to hot deformation parameters. In conventional manufacturing routes, numerous trial and error steps must be taken to achieve the suitable processing parameters. Therefore, it is very useful to develop an effective model to predict deformation behaviors.
Alloy GH742 is used as sample to study the hot workability of highly strengthened superalloys. First, the material database was set up using compression tests with deformation temperature from 1050 to1125℃, engineering strain from 0.1 to 0.7; and the true strain rates of 0.1, 0.01 and 0.001S-1. The microstructure of deformed samples was examined by optical and electrical microscope.
Empirical metallurgical equations were developed to describe the relationship between flow stress deformation ductility, dynamic recrystallization(DRX) and deformation parameters. The equations should not only describe the experimental results very well, but also reflect the phenomena occurring during hot deformation.
.
These equations and simulated results are as follows
(1) The peak flow stress vs. parameters
Fig.2 gives the experimental relationship of peak flow stress with deformation temperature and true strain rate. The following is the simulated equation.
where A is a constant, n is a parameter related to the true strain rate, Q is the deformation activation energy and R is the gas constant.
The above equation shows that lnand ln,lnQ and lnT have a linear relationship as same as the experimental results of Fig.2. Therefore, the parameter n, Q and A can be obtained from Fig. 2. N=0.159 Q=2177564.11 A=1.824. Fig.3 shows the simulated results, it predicts the experimental results very well.
Fig. 2 The relationship of peak flow stress with deformation temperature and true strain rate.
(a) experimental results (b) calculated results
Fig 3. The experimental and calculated peak flow stress
(2) Hot defamation ductility vs. parameters
Simulated equation:
The effects of true strain rate on the ductility are reflected in parameters A,f1, and f2. The equations obtained by statistical fit are as follows:
Fig.4 shows the calculated and experimental hot deformation ductility. It is seen that the model predicted the hot deformation ductility very well.
(a) experimental results (b) caculated results
Fig.4 The experimental and calculated hot deformation ductility
(3) Percent dynamic recystallization (DRX) vs. parameters
Simulated equation:
where P1 is the slope rate that describes the gradient of cures and ε50 is the critical strain which indicates 50% DRX occurred during hot deformation.
The equations obtained by statistical fit of experimental data.
P1 = 27.60691-0.05203×T+2.46E5×T2
50 = 6291.5-10.46586×T+0.00436×T2
Fig.5 presents the calculated and experimental percent dynamic recrystallization, it shows the equations describe the relationship between percent DRX and deformation parameters very well.
(a) experimental results (b)calculated results
Fig.5 The experimental and calculated percent DRX
Constitutive Equation Model of Superalloy Inconel 718 for High Speed Counter-blow Hammer Forging
Inconel 718 has been most widely used as turbine discs because of its good performance and low cost. However, it is difficult to control microstructure of forging during deformation processing because of complicated phase composition.
In the past decade, some numerical models were set up to predict the microstructure evolution during deformation. Most of them were based on isothermal constant speed pancake forging process.
For economical view, high-speed counter-blow hammer forging is widely adopted because of very efficiency. Therefore, our work focuses on the simulation of high-speed multi-pass deformation process, the investigation of the deformation behavior influenced by hot deformation parameters, and development of the constitutive equation reflecting the flow rule of high-speed counter-blow hammer forging process.
Experiment Method
Counter-blow hammer forging process is different to isothermal pancake forging process, more than 30 deformation passes should be needed to get finished forging. Within the interval time, static recovery and recrystallization will be occurred. Since more than 70% strain should be completed in initial several passes, the 3-passes and 5 passes deformation schedule have been adopted to simulate the practice deformation history. The distribution of engineering strain for 3 passed and 5 passes is 0.4+0.2+0.1 (0.7 in total) and 0.25+0.15+0.15+0.1+0.05 (0.7 in total) respectively. The compress testing was carried out with High Speed Testing Machine. The testing parameters were adopted as follows.
Deformation temp.: C – Cwith an interval of C.
Strain rate:10/s.
Interval times:5 second
Heating hold time: 30 min.
Simplification of the stress-strain Curve
3-passes and 5-passes true stress-true strain curves, which reflects the thermo-mechanical history of compression specimen, are shown in Fig.6. These curves are difficult to express in mathematics. In consideration of the same interval time between two passes, the effect of interval time on the flow stress is constant. Hence, the effect can be synthesized as one constant state variable. Fig.6 shows there are work hardening and recrystallized softening in the recorded stress-strain curve as a whole, which is similar with the stress-strain curve of conventional isothermal constant speed forging process. Therefore, the multi-pass stress-strain curve can be simplified and quantified as the only single pass forging process as showed in Fig.7. It is easy to describe the stress-strain curve mathematically.
Fig.6 True stress-true strain curve
Fig.7 Simplification figure of multi-pass deformation true stress-strain curve (n=5)
Constitutive Equation
The relationship between the flow stress and the strain can be outlined by following formula.
=A×n×exp(f())
where n is work hardening rate, A is a constant.
The relationship between the flow stress and deformation temperature can be described as follows:
T = B0+ B1×exp[(B2-T0)/B3]
where T is deformation temperature, Bo,B1,B2 and B3 are constants.
Flow stress is the function of strain and deformation temperature, at the constant strain rate with 10/s. The synthesized constitutive equation is as follows.
=A×n×exp(f())×{B0+ B1×exp[(B2-T0)/B3]}
The simulation curves, which are calculated from constitutive equation, are consistent with experiment value in Fig.8. The simplified schedule of high-speed counter-blow hammer forging process of superalloy In718 is reasonable and feasible.
The microstructure model have been developed as follows:
D = Dmin×Fmin+ Dman×Fman
D = Co×exp(ε)×exp[C2/(Z+C3)×t×(C4+C5×t + C6×t2)]
D — the average grain size without holding process
Dmin — the recrystallization grain size
Fmin — the recrystallization fraction
Dman — the unrecrystallization grain size
Fman — the unrecrystallization fraction
D — the increment of the grain size owing to the hold process after
deformation
C0, C2, C3, C5 and C6 are constants.
Fig.8 Simulation results of stress-strain curve (n=5)
Fig.9 gives the typical examples of simulation results of the grain field of final forged disc, it shows the calculated microstructure by numerical model is similar to the actual experimental results of disc.
Fig.9 Simulation results of the grain field of final forged disc
Above mentioned are several examples of CAE application in superalloy casting and deforming processes. The influence of computer modeling to simulate casting and deforming processes is revolutionizing process development in the superalloy field. It will help improve affordability of superalloy in the competitive market.
References
- J.M.Zhang, L.Z.Ma, “Superalloy 718,625,706 and various Derivatives” Edited by E.A.Loria. The Minerals, Metals and Materials Society 1997, P183-192.
- Z.Long, J.Zhang, “Advanced Technologies for Superalloy Affordability” Edited by K.M.Chang, S.K.Srivastava,2000. P187-196.
- David Furrer and Hans Fecht, Ni-Based Superalloys for Turbine Discs, JOM, January 1999,PP14-18
- M.J.Goulette, The Future Costs Less-High Temperature Materials From an Aeroengine Perspective, Superalloys 1996, eds., R.D.Kissinger et al. (Warendale:TMS 1996) , PP3-6
- D.Lasalmonie, et al., A Computer assisted Design and Modeling Tool For Superalloy Powder Precision Modeling, Superalloys 1996, (Warendale:TMS 1996) PP737-743
- .M. Zhang, Z.Y.Gao, J.Y.Zhuang, Q.Deng, and P.Janschek, J. Mater. Processing Technol., 1997, vol.70, pp252-257
- Z. Long, J.Zhuang, B.Deng, P.Lin, Z.Zong, The Hot Deforamtion Modeling of As-Cast High-Strengthened superalloys, Advanced Technologies for Superalloy Affordability, eds. K-M Chang, S.K.Srivastava, D.U.Furrer and K.R.Bain, (Nashville, March TMS,1997) PP187-196.
- Z. Long, J.Zhuang, B.Deng, P.Lin, Z.Zong, Advanced Technologies for Superalloy Affordability, eds. K-M Chang, S.K.Srivastava, D.U.Furrer and K.R.Bain, (Nashville, March TMS,1997) PP287-297.
- F. Hediger and N. Hofmann. Proc.of the 5th Intern. Conference in Modeling of Casting, Welding and advanced Solidification Processes, Davos, Switzerland, sept. 16-21, 1990, eds, M.Rappaz et al., Warendale:TMS 1991, p.611-619.
1