In my years of working with machine tool castings, I have encountered numerous challenges related to internal stresses. These stresses arise primarily due to uneven cooling during the solidification process of cast iron components, which is common in machine tool casting. The variation in wall thickness leads to differential cooling rates, inducing significant residual stresses. Additionally, machining operations introduce new cutting stresses. The presence of these stresses can cause distortion, cracking, and ultimately compromise the precision and stability of machine tools. Therefore, effective stress relief methods are critical for ensuring the quality and longevity of machine tool castings.
Traditionally, several methods have been employed to alleviate internal stresses in machine tool castings, including natural aging, artificial aging (such as oven heating), and mechanical aging. Among these, oven heating with controlled temperature cycles is widely used for its efficiency in batch production. The typical temperature control curve involves heating, soaking, and cooling phases, as illustrated below:
This process requires sophisticated thermal control instrumentation. However, in recent decades, vibration stress relief (VSR) has emerged as an alternative technique. Initially explored in the Soviet Union, Japan, and the United States, VSR has gained attention for its potential to reduce stress without thermal input. Over the past three years, we have actively investigated VSR for machine tool castings, particularly applying it to components like the worktable of a crankshaft grinder. This article shares my insights, experimental findings, and theoretical perspectives on why vibration can effectively eliminate internal stresses in machine tool castings.
To understand the mechanism, we conducted experiments using a classic stress frame specimen. This stress frame, commonly used in stress analysis, consists of a central rod with a larger cross-section flanked by two side rods with smaller cross-sections, connected by horizontal beams. When cast, the side rods solidify first, leaving the central rod under tension during cooling, while the side rods experience compression and the beams undergo bending. This setup simulates the stress state in complex machine tool castings. By sawing the central rod and measuring the resulting gap, we can infer the internal stress magnitude. For our tests, we divided stress frames from the same melt into three groups: Group 1 (as-cast, no treatment), Group 2 (vibrated with insufficient force), and Group 3 (vibrated with adequate force). The gap widths were measured to assess stress relief.
The internal stress distribution in the stress frame can be quantified through mechanical calculations. Assuming the central rod has a cross-sectional area $A_c$, the side rods have area $A_s$, and the beams have length $L$ and moment of inertia $I$, we derive the following equations based on force equilibrium and compatibility. Let $\sigma_c$ be the tensile stress in the central rod, $\sigma_s$ the compressive stress in the side rods, and $\delta$ the elongation of the central rod upon release. The total deformation $\Delta$ is given by:
$$ \Delta = \delta_c + \delta_b + \delta_s $$
where $\delta_c$ is the elongation due to $\sigma_c$, $\delta_b$ is the beam deflection due to bending, and $\delta_s$ is the shortening of side rods due to load transfer. Using Hooke’s law and beam theory, we express:
$$ \delta_c = \frac{\sigma_c L}{E}, \quad \delta_b = \frac{FL^3}{3EI}, \quad \delta_s = \frac{\sigma_s L}{E} $$
Here, $E$ is the Young’s modulus of the cast iron, and $F$ is the force acting on the beams. For simplicity, neglecting shear effects, the relationship simplifies to:
$$ \sigma_c = \frac{E \Delta}{L} \cdot \frac{1}{1 + \frac{A_c}{2A_s} + \frac{L^2 A_c}{3I}} $$
Similarly, the compressive stress in the side rods is:
$$ \sigma_s = -\frac{A_c}{2A_s} \sigma_c $$
And the maximum bending stress in the beams, $\sigma_b$, is:
$$ \sigma_b = \frac{M y}{I} = \frac{F L y}{2I} $$
where $M$ is the bending moment and $y$ is the distance from the neutral axis. For our stress frame, with dimensions: central rod width $b_c = 50\, \text{mm}$, side rod width $b_s = 20\, \text{mm}$, beam length $L = 200\, \text{mm}$, and beam height $h = 30\, \text{mm}$, we calculate the stresses. The results from Group 1 are summarized in Table 1, showing average stresses before treatment.
| Frame No. | Measured Gap $\delta$ (mm) | Central Rod Tensile Stress $\sigma_c$ (MPa) | Side Rod Compressive Stress $\sigma_s$ (MPa) | Beam Bending Stress $\sigma_b$ (MPa) |
|---|---|---|---|---|
| 1 | 0.15 | 45.2 | -22.6 | 67.8 |
| 2 | 0.18 | 54.3 | -27.1 | 81.4 |
| 3 | 0.16 | 48.1 | -24.1 | 72.2 |
| Average | 0.163 | 49.2 | -24.6 | 73.8 |
These values indicate that bending stresses dominate, which is typical in machine tool casting geometries. After vibration treatment, Group 3 showed reduced gap widths below 0.05 mm, implying significant stress relief, whereas Group 2 showed minimal change. This underscores the importance of sufficient vibratory force.
To delve deeper, we performed tensile tests on gray cast iron (Grade 2) specimens, simulating the material used in machine tool castings. The stress-strain ($\sigma$-$\varepsilon$) curve, recorded with high sensitivity, revealed key insights. As shown in Figure 1 (not included here, but described), upon loading to near ultimate strength $\sigma_u \approx 200\, \text{MPa}$, unloading followed a path that left permanent strain $\varepsilon_p$. Reloading showed a higher elastic modulus $E’$, indicating strain hardening. The yield stress $\sigma_y$ for as-cast iron is approximately 150 MPa, with corresponding strain $\varepsilon_y = \sigma_y / E$. After plastic deformation, $E’$ increases significantly, enhancing stiffness. This behavior is crucial for understanding VSR.
Based on these experiments, I propose that vibration eliminates internal stresses in machine tool castings through a mechanism of superimposed dynamic strain. When a casting is subjected to vibratory forces at resonant frequencies, the alternating stresses add to the existing residual stresses. If the combined stress exceeds the yield stress $\sigma_y$ at any point, localized plastic deformation occurs, relieving the stress. Mathematically, for a point with initial residual stress $\sigma_r$, the total stress during vibration is:
$$ \sigma_{\text{total}} = \sigma_r + \sigma_v \sin(\omega t) $$
where $\sigma_v$ is the vibratory stress amplitude, $\omega$ is the angular frequency, and $t$ is time. Plastic deformation ensues when:
$$ |\sigma_r + \sigma_v| \geq \sigma_y $$
This is more likely in regions with high $\sigma_r$, such as thick sections in machine tool castings. Over multiple cycles (e.g., 3 minutes at 50 Hz, totaling 9,000 cycles), cumulative micro-yielding reduces $\sigma_r$ until $\sigma_{\text{total}} < \sigma_y$. This process differs from fatigue (which requires millions of cycles) or creep (which requires prolonged time), aligning with our observations.
The effectiveness of VSR depends on several factors. First, the vibratory force must be adequate to generate sufficient $\sigma_v$. We found that for our stress frames, a force below 500 N had no effect, whereas 2000 N induced clear relief. Second, the excitation direction should align with the principal stress directions. In our tests, vibration along the central rod axis (tensile direction) was most effective. Third, resonance tuning maximizes energy transfer; we used accelerometers to identify natural frequencies around 50-100 Hz for typical machine tool castings.
To quantify the stress relief, we can model the strain accumulation. Let $\varepsilon_e$ be the elastic strain and $\varepsilon_p$ the plastic strain. The total strain during vibration is:
$$ \varepsilon = \varepsilon_e + \varepsilon_p = \frac{\sigma}{E} + \int_{0}^{t} \dot{\varepsilon}_p \, dt $$
where $\dot{\varepsilon}_p$ is the plastic strain rate, governed by a yield criterion. For cast iron, a simplified model gives:
$$ \dot{\varepsilon}_p = C \left( |\sigma| – \sigma_y \right)^n $$
with $C$ and $n$ as material constants. Integration over vibration cycles shows exponential decay of $\sigma_r$, similar to relaxation phenomena. This explains why short-duration VSR can be effective for machine tool castings.
Compared to thermal methods, VSR offers advantages. Heating to 500-600°C reduces yield stress to about 50 MPa, allowing stress relief through plasticity. However, this consumes energy and may cause distortion. VSR operates at room temperature, saving energy and time. Moreover, as observed in our tensile tests, post-yield elastic modulus increase enhances rigidity, beneficial for machine tool casting performance. Table 2 contrasts the methods.
| Method | Temperature | Duration | Energy Use | Stress Reduction | Effect on Stiffness |
|---|---|---|---|---|---|
| Natural Aging | Ambient | Months | Low | Slow, partial | Unchanged |
| Oven Heating | 500-600°C | Hours | High | Significant | May decrease |
| Vibration (VSR) | Ambient | Minutes | Low | Significant | Increases |
Our practical application on crankshaft grinder worktables demonstrated VSR’s efficacy. We divided worktables into groups: one with VSR, one without, and one with partial treatment. The VSR-treated group achieved stable machining accuracy in one pass, with distortion below 0.01 mm, whereas untreated ones required multiple passes and still showed variability. This confirms that VSR can stabilize machine tool castings for precision machining.
In conclusion, vibration stress relief is a viable technique for internal stress reduction in machine tool castings. Its principle hinges on dynamic superposition inducing localized yielding, which is efficient and non-thermal. Key parameters include vibratory force, frequency resonance, and directionality. Further research could optimize these parameters for complex casting geometries. As the demand for high-precision machine tools grows, VSR presents a sustainable alternative to traditional methods, enhancing the quality and performance of machine tool castings.
To summarize the mathematical framework, the stress relief $\Delta \sigma_r$ after vibration time $T$ can be estimated as:
$$ \Delta \sigma_r = \sigma_r^0 \left[ 1 – \exp\left( -\frac{T}{\tau} \right) \right] $$
where $\sigma_r^0$ is the initial residual stress and $\tau$ is a time constant dependent on material properties and vibration intensity. For gray cast iron in machine tool castings, $\tau$ is typically on the order of minutes under resonant conditions. This exponential decay model fits our experimental data well.
Ultimately, the integration of VSR into manufacturing processes for machine tool castings can lead to improved dimensional stability, reduced scrap rates, and lower energy consumption. As we continue to refine this technology, it holds promise for broader adoption in the foundry and machining industries, ensuring that machine tool castings meet ever-tighter tolerances and reliability standards.