In the production of heavy machine tool castings using split-box molding, the bottom-return rain gating system is widely adopted due to its significant advantages in ensuring smooth filling and uniform temperature distribution. However, defects in such large-scale machine tool castings can lead to substantial economic losses. Therefore, developing a rational gating design that enables rapid and stable mold filling is crucial for producing high-quality machine tool castings consistently. This study focuses on addressing quality issues in heavy machine tool castings by establishing a comprehensive theoretical calculation method for the bottom-return rain gating system, providing experimental data and process improvements to reduce or eliminate defects caused by improper design, thereby enhancing economic efficiency in casting production.
We conducted hydraulic simulation experiments to investigate the bottom-return rain gating system, measuring physical parameters such as gate velocity, pressure, and flow distribution during mold filling. By observing flow patterns in various sections, we analyzed the filling characteristics theoretically, considering the effects of bottom gating, and explored the formation mechanisms of several casting defects. Our findings reveal conclusions distinctly different from top-gating systems, and we implemented computer-aided design for the bottom-return rain gating system in bed-type machine tool castings.
The foundation of our water simulation experiments lies in the mechanical similarity theory of flow. For the bottom-return rain gating system, gravitational and resistive forces dominate the fluid dynamics during pouring. Thus, we ensured satisfaction of two similarity criteria: Froude similarity and Reynolds similarity. The prototype was selected based on surveys of actual production processes for machine tool castings. We chose a scale factor of 1:1, meaning the model size equals the prototype, to accurately reflect the flow behavior of molten iron in the mold. This allows direct application of experimental parameters to production without conversion. The structural dimensions of the two-row inner gate bottom-return rain gating system model are summarized in Table 1.
| Parameter | Value |
|---|---|
| Model length (mm) | 1200 |
| Model width (mm) | 800 |
| Model height (mm) | 600 |
| Pouring cup orifice diameter (mm) | φ20, φ25, φ30, φ35, φ40, φ45 |
| Sprue diameter (mm) | φ30 |
| Runner cross-sectional area (mm²) | 30 × 40 |
| Branch sprue cross-sectional area (mm²) | 25 × 30 |
| Branch runner cross-sectional area (mm²) | 20 × 25 |
| Inner gate cross-sectional area (mm²) | 15 × 20 |
To simultaneously record the flow characteristics at each inner gate, we designed a multi-point measurement setup for flow rate. Contacts were placed at different heights in the boxes corresponding to the inner gates, and the time for the liquid to reach each contact was recorded using an optical oscillograph, generating height-time curves. Since each box has a constant cross-section, we derived the flow rate-time relationship for each inner gate and the start time of filling for each box. Pressure measurement involved pressure tubes connected to the inner gates, with contacts inside the tubes to record pressure variations over time, as shown in Figure 1. Velocity at the inner gates was measured using a laser Doppler velocimeter, which utilizes the optical Doppler effect to determine fluid velocity by analyzing the frequency shift of scattered light from particles in the flow. Instantaneous velocity, average velocity, turbulence intensity, and velocity distribution were output via a microprocessor.
The flow distribution characteristics of the inner gates in the bottom-return rain gating system were analyzed under various pouring conditions. Unlike top-gating systems, which often exhibit uneven flow distribution, our experiments showed that the bottom-return system provides relatively uniform flow allocation. The flow distribution ratio for each inner gate was measured under different pouring cup liquid levels (H = 100 mm, 150 mm, 200 mm, 250 mm, 300 mm) and orifice diameters (d = 20 mm to 45 mm). Results indicate that inner gates farther from the branch sprue bottom tend to have slightly higher flow rates due to “superposition” flow in the branch runner. When the pouring cup level is below 200 mm, kinetic energy loss can lead to a “middle-heavy, ends-light” distribution, but at higher levels, flow distribution stabilizes. Table 2 summarizes the flow distribution ratios under selected conditions.
| Inner Gate Number | Flow Distribution Ratio (%) |
|---|---|
| 1 | 15.2 |
| 2 | 16.8 |
| 3 | 17.5 |
| 4 | 18.1 |
| 5 | 17.9 |
| 6 | 16.5 |
The pressure distribution at the bottom of each inner gate was found to correlate with flow distribution. During filling, the branch runner and inner gates remain full, acting as a communicating vessel. Pressure values increase with pouring cup level and vary with orifice diameter. At lower levels (H < 200 mm), larger orifice diameters result in lower pressure due to increased frictional losses along the sprue wall. At higher levels, pressure rises with orifice diameter. The pressure-time curve shows a sudden spike at the start of filling due to jetting, then stabilizes. The pressure distribution can be expressed theoretically using Bernoulli’s equation modified for bottom gating:
$$P_i = \rho g H_i – \frac{1}{2} \rho v_i^2 – \sum \Delta h_{loss,i}$$
where \(P_i\) is the pressure at inner gate \(i\), \(\rho\) is fluid density, \(g\) is gravitational acceleration, \(H_i\) is the effective head, \(v_i\) is velocity, and \(\sum \Delta h_{loss,i}\) is the total head loss. For machine tool castings, this ensures stable pressure during filling.
Velocity distribution in the inner gates was measured using laser Doppler velocimetry. Under constant pouring cup level (H = 200 mm) and orifice diameter (d = 30 mm), the liquid level rise in the mold cavity over time was linear, indicating constant velocity at each inner gate throughout filling. This is a key advantage of the bottom-return rain gating system for machine tool castings, as it prevents turbulent flow and promotes uniformity. The relationship between velocity and pouring parameters is given by:
$$v_i = \sqrt{\frac{2gH_{eff,i}}{1 + \sum \zeta_i}}$$
where \(H_{eff,i}\) is the effective head considering back pressure, and \(\sum \zeta_i\) is the sum of resistance coefficients. Experimental velocity values are shown in Table 3.
| Inner Gate Number | H = 150 mm, d = 25 mm | H = 200 mm, d = 30 mm | H = 250 mm, d = 35 mm |
|---|---|---|---|
| 1 | 1.2 | 1.5 | 1.8 |
| 2 | 1.3 | 1.6 | 1.9 |
| 3 | 1.4 | 1.7 | 2.0 |
| 4 | 1.4 | 1.7 | 2.0 |
| 5 | 1.3 | 1.6 | 1.9 |
| 6 | 1.2 | 1.5 | 1.8 |
Flow visualization using transparent acrylic models revealed critical phenomena affecting machine tool castings quality. An “impact mixing zone” forms where the liquid stream falls into the branch sprue, creating turbulence and gas entrainment. Bubbles generated here can enter the mold cavity through the first inner gate near the branch runner, leading to gas porosity in machine tool castings. To mitigate this, we observed that a “dead zone” at the junction of branch sprue and runner traps gas, forming a “V-shaped” air region. By adding a degassing structure with vent holes, such as a “gas collection pocket,” bubble entry was significantly reduced. Jetting phenomena occurred when liquid rapidly entered the inner gates due to sudden area contraction, causing defects like slag inclusion and sand erosion. Installing tapered slag traps with vent holes at the branch runner end effectively eliminated jetting and improved slag removal. Simulating slag movement with polystyrene particles showed that most slag accumulates at the runner end and mold top, and centrifugal slag traps enhanced removal efficiency.
Numerical analysis of the bottom-return rain gating system was based on hydraulic principles. Optimizing pouring time is essential for designing gating systems for machine tool castings. We formulated it as a constrained nonlinear programming problem to minimize pouring time \(T\) subject to constraints on mold filling velocity and time limits:
$$\text{Minimize } T$$
$$\text{Subject to: } V_{min} \leq \frac{H_c}{T} \leq V_{max}$$
$$T_{min} \leq T \leq T_{max}$$
where \(H_c\) is casting height, \(V_{min}\) and \(V_{max}\) are allowable filling velocities, and \(T_{min}\) and \(T_{max}\) are time bounds. Using the complex method, we solved for optimal pouring time \(T^*\). The mathematical model applies Bernoulli’s equation between the pouring cup and each inner gate, accounting for head losses:
$$H_{eff,i} = \frac{H_i^2}{2H_c} \left(1 – \left(1 – \frac{h_i}{H_c}\right)^2\right)$$
for bottom gating, where \(h_i\) is the liquid height in the cavity. Head losses include frictional and local losses:
$$\sum \Delta h_{loss,i} = \sum \lambda_j \frac{L_j}{D_j} \frac{v_j^2}{2g} + \sum \zeta_k \frac{v_k^2}{2g}$$
where \(\lambda_j\) are friction coefficients, \(L_j\) and \(D_j\) are lengths and hydraulic diameters, \(\zeta_k\) are local loss coefficients, and \(v_j\), \(v_k\) are velocities in sections. Experimental data were used to calibrate resistance coefficients. The inner gate velocity is then:
$$v_i = \sqrt{\frac{2gH_{eff,i}}{1 + \sum \lambda_j \frac{L_j}{D_j} + \sum \zeta_k}}$$
Given the weight \(W\) of machine tool castings, density \(\rho\), optimal pouring time \(T^*\), and velocity \(v_i\), the inner gate area \(A_i\) is calculated as:
$$A_i = \frac{W}{\rho T^* \sum v_i}$$
For uniform wall thickness bed-type machine tool castings, with \(n\) inner gates per row, the area per gate simplifies to:
$$A_i = \frac{W}{n \rho T^* v_i}$$
Based on production experience, gating ratios for open or partially choked systems are typically:
$$\sum A_{sprue} : \sum A_{runner} : \sum A_{branch\ runner} : \sum A_{inner\ gates} = 1 : 1.5 : 2 : 3$$
We implemented computer-aided design for the bottom-return rain gating system, integrating data processing, pouring time optimization, gating calculations, and automatic drawing generation. The CAD program, developed in FORTRAN, includes modules for inputting casting parameters, calculating head losses and flow coefficients, optimizing pouring time, determining gating dimensions, and outputting process cards. This system enhances design accuracy and efficiency for machine tool castings production.
In conclusion, hydraulic simulation experiments are effective for optimizing gating systems in complex machine tool castings. The bottom-return rain gating system ensures non-turbulent filling, with uniform flow distribution, constant gate velocity, and stable pressure, benefiting bed-type machine tool castings by reducing stresses and improving hardness uniformity. Key improvements include degassing structures to prevent gas porosity, centrifugal slag traps to eliminate inclusions and jetting, and maintaining pouring cup levels between 200-300 mm for stable filling. Numerical modeling and CAD integration provide reliable design tools. Our study underscores the importance of considering back pressure in non-top-gating systems and offers practical solutions for enhancing the quality of heavy machine tool castings.
Further research could explore the effects of molten metal properties, such as viscosity and surface tension, on flow behavior, and extend the model to multi-row gating systems for larger machine tool castings. The integration of real-time monitoring during pouring may also advance process control in foundries.