An integrated systems approach to understanding distortion and residual stress during thermal processing: design for heat treating
Heat treatment processes are used to develop the desired mechanical properties for steels. Unfortunately, heat treatment, especially quenching, can cause distortion. Failure to meet geometry specifications can result in extensive rework or rejection of the parts. A series of quenching simulations, using DANTE, have been conducted on an AISI 4140 steel Navy C-ring distortion coupon and a WPI designed plate with a hole to determine the effects of quenching process parameters including part geometry, agitation during quenching, and quench start temperatures on distortion. The heat transfer coefficients (HTC) of the quenchant with selected pump speeds were measured by CHTE quench probe system, which is the key input for heat treatment simulation. The maximum HTC of the quenching oil was increased from 2350 W/m2K to 2666 W/m2K with higher pump speed. Quenching experiments were also conducted. It was found that the experimental measured gap opening of the standard Navy C-rings increased from 0.307mm without agitation to 0.536mm at a high agitation. Quench start temperature does not have a significant effect on the gap opening. The experimental results showed good agreement with simulation results. The important processing parameter identification was conducted using design of experiments (DoE) coupled with analysis of variance (ANOVA). The effect of processing parameters in decreasing order of importance were determined to be: quenchant type, part geometry, agitation speed, quenching orientation, quenchant temperature, immersion rates, and quench starts temperature. Based on the simulation and experimental results, it was found that the two most import parameters are: 1. The part geometry and size (product design) 2. The temperature dependent heat transfer coefficients between the part and the quenchant (process design) The coupling of these product and process parameters is necessary to apply the systems analysis that must be accomplished to understand the interaction between the part design and process design parameters. This coupling can be accomplished by locally applying the well-known Biot number. Bi (T) = h(T) * L / k(T) Where h(T) = film coefficient or convective heat transfer coefficient [W/m2*K]. LC = characteristic length, which is generally described as the volume of the body divided by the surface area of the body [m]. k(T) = thermal conductivity of the body [W/m*k] The concept of a local Biot number is introduced to quantify the local variations of part size, geometry and heat transfer coefficient. First, a large Bi indicates large temperature gradients within the part. Second, large local (geometry dependent) variations in Bi number will lead to large lateral temperature gradients. Therefore, variations in local Bi can lead to large temperature gradients and therefore high stress during quenching and finally distortion. This local Bi concept can be used in a systems approach to designing a part and the quenching system. This systems approach can be designated as design for heat treating.