Reviewing aluminum gating systems: new technology has allowed the testing of several mold filling notions the metalcasting industry has long held dear.
The filling patterns of various aluminum gating systems for sand casting recently were investigated by observing both water modeling and molten aluminum via a new video capturing technique, allowing each systems' coefficient of discharge to be measured. Non-pressurized systems were shown to produce lower metal velocities and minimize the possibility of inclusion and oxide entrance. In addition to the lower velocities, design enhancements, like positioning the gates above runners, contributed to minimizing the defects.
The measured coefficients of discharge showed that the distance between gates and sprues, gating ratios, fluid velocity, area-to-perimeter ratios, gate width-to-length ratios, and runner length were critical for producing sound castings. Generally, greater gate-to-sprue and runner-to-sprue ratios and smaller ratios of thickness to width will result in higher coefficients of discharge and increased flow rates.
Following is a look at the results of the water model testing performed on seven different aluminum gating systems.
7 Systems Analyzed
Seven common aluminum gating systems, both pressurized and non-pressurized, were examined in the modeling experiments. The gating systems are shown in Fig. 1, and the design details are listed in Table 1.
Table 1. Design Details of Gating Systems (mm) System Sprue:Runner:Gate * Sprue Diameter Well (S:R:G) Top Bottom Height Depth 1 1:0.9:0.8 38 25 60 50 2 1:0.7:0.5 38 25 52 50 3 1.6:2:4 38 25 36 50 3A 1.6:2:4 38 25 36 50 4 1.6:4:4 38 25 50 50 5 1:2.3:4.5 38 25 48 50 5A 1:2.3:4.5 38 25 48 50 System Runner Gate Length Width Width ** No. Length Width 1 15 30 9 2 28 7 2 13 26 9 2 22 5.6 3 36 18 6 2 30 21 3A 36 18 6 2 42 15 4 50 25 6 2 30 21 5 47 24 12 2 56 20 5A 47 24 6 7 32 10 * Area ratio of sprue:runner(s):gate(s) ** Runner thickness at end
Pressurized (System 1) 1:0.9:0.8--Turbulent flow, incomplete runner filling, vena contracta (effective choke) at the runner/gate junctions, and splashing of liquid were apparent throughout the filling process. In addition, a molten aluminum test verified the vena contracta in sand casting.
Pressurized (System 2) 1:0.7:0.5-System 2 exhibited a similar flow pattern to System 1. Due to smaller ratios in comparison to the first system, the gate velocity was higher. Vena contracta and premature fluid exit from the gates were observed.
Fluid motion in pressurized systems has a particular pattern. After the flow changes direction from vertical to horizontal (sprue to runner), the fluid is turbulent and contains bubbles. Since the ratio gradually decreases in a pressurized system, the velocity increases progressively and flow becomes increasingly agitated. By designing the choke at the gate(s), the system is filled backwards from the gate(s) and, due to the high velocity that results from locating the gate(s) below the runners, fluid enters the mold before completely filling the system. The turbulence and high velocity facilitate inclusions and air pocket entry into the mold cavity.
[FIGURE 1 OMITTED]
The vena contracta, forming as a result of the larger cross-section of the runner compared to the gates and sharp changes of direction (which is intensified in the water model), is observed where the fluid exits the gates. Both the water modeling and aluminum casting showed the second gate yielded lower vena contracta, which is associated with back pressure. In addition, at the point of vena contracta, the contact of the fluid stream with the gating system wall is lost, which can be a source of air entrainment.
Non-Pressurized (System 3) 1.6:2:4--The metal flow was less turbulent in System 3, and the runner filled completely before the gates. A new phenomenon was observed in the system--vortex formation during mold filling. The vortex was studied in the aluminum casting from both the front and back views. As seen in Fig. 2, the liquid contacted the gate walls, surpassed the gate's peripheral area, formed a vortex, and finally entered the gate through a stirring motion. This phenomenon was likely the result of the high velocity in the runner and high ratio of gate width to length.
[FIGURE 2 OMITTED]
The video recordings of the water modeling and molten aluminum experiments showed that the liquid initially contacted the runner walls at high velocity and was guided past the entrances of the gates. When it finally entered the gates, the flow surpassed the gate's peripheral area and formed streams due to the gravity and velocity reduction. In complete filling of the gates and formation of a low-pressure zone could result in air aspiration and more double oxide film formation. The probability of vortex formation can be minimized by decreasing the gate-width-to-length ratio. The fluid velocity in the runner also can be reduced by appropriate design and installation of other features, such as a filter.
Non-Pressurized (System 3A) 1.6:2:4--This system was designed to overcome the vortex formation. It was shown that reduction of the length-to-thickness ratio can eliminate the vortex.
Non-Pressurized (System 4) 1.6:4:4--In System 4, the fluid demonstrated the same behavior as in the other non-pressurized systems. However, the system presented the advantage of inducing less turbulence.
Non-Pressurized (System 5) 1:2.3:4.5--This system was designed to yield the critical gate velocity of 0.5 m/s, which allows mold filling with the minimum acceptable turbulence. A vortex also was created in this system; however, due to the extended length of the gates, the vortex was unstable and eliminated instantly.
Non-Pressurized (System 5A) 1:2.3:4.5--In this alternative to System 5, reducing thickness and increasing width were shown to be effective in eliminating the vortex. (Due to the improper tapering of the runner, the fluid exit from the gates was not simultaneous.) The system also raised questions about the number of gates used, which depends on the geometry of the casting and design preferences. For instance, numerous gates lead to better directional solidification, while long runners lead to lower velocities and higher discharge coefficient values that can result in incomplete mold filling.
In aluminum gating systems, the fluid velocity at the bottom of the sprue results in the runner being initially filled with turbulent metal. However, in non-pressurized systems, the runner cross-section is larger, which allows the fluid to become more stable and tranquil. By positioning the gates over the runner and in the cope, the runner is filled initially before fluid enters the gate. Inclusions and oxide particles are mostly trapped at the end of the runner and remain attached to the upper surface. Hence, the chance of contaminated metal entering the mold cavity becomes reduced. By designing a properly tapered runner (not a stepped runner, since the melt tends to be affected turbulently, bouncing at each step), the metal enters the mold concurrently through all the gates.
The measured gate velocities, actual filled areas and flow rates are presented in Tables 2-4. Theoretical gate velocity was determined assuming complete filling of the gate area, while the actual gate velocity was determined based on the flow rate and proportion of area filled, which was measured by reviewing the recorded frames in the measured time. Using the actual velocity, the velocity coefficient of discharge was calculated. The same concept was used for measuring the areal coefficient of discharge (CdA).
Table 2. Gate Velocity Measurement and Velocity Coefficient of Discharge System No. Sprue:Runner:Gate R/S G/S [(A/P).sub.G] [(A/P).sub.R] 1 1: 0.9: 0.8 0.9 0.8 2.8 5 2 1:0.7: 0.5 0.7 0.5 2.2 4.7 3 1.6: 2: 4 1.28 2.56 6.2 6 3A 1.6: 2:4 1.28 2.56 5.5 6 4 1.6: 4: 4 2.56 2.56 6.2 8.3 5 1: 2.3: 4.5 2.27 4.5 7.2 8 5A 1: 2.3: 4.5 2.27 4.5 3.8 8 System No. [V.sub.Gt] [V.sub.G1] [V.sub.G2] [CdV.sub.1] [CdV.sub.2] 1 2.85 1.83 1.66 0.64 0.58 2 4.54 1.98 1.7 0.44 0.37 3 0.88 0.59 0.65 0.67 0.74 3A 0.88 0.58 0.6 0.66 0.68 4 0.88 0.61 0.69 0.69 0.78 5 0.5 0.45 0.48 0.9 0.96 5A 0.5 0.45 0.46 * 0.9 0.92 System No. VG ave CdV 1 1.75 0.61 2 1.84 0.4 3 0.62 0.71 3A 0.59 0.67 4 0.65 0.74 5 0.47 0.93 5A 0.4 0.8 * Us values for the remaining gates were 0.45, 0.4, 0.35, 0.36 and 0.31 A: area ([m.sup.2]), P: perimeter (m), [V.sub.Gt]: theoretical gate velocity (m/s), [V.sub.G]: real gate velocity (m/s), CdV: velocity coefficient of discharge (average of [CdV.sub.1] and [CdV.sub.2]), [V.sub.G] Ave: average of gate velocities (m/s) Table 3. Filled Area of Gates and Areal Coefficient of Discharge System No. Sprue:Runner:Gate G/S [(T/W).sub.g] [A.sub.Gt] 1 1: 0.9: 0.8 0.8 0.25 196 2 1: 0.7: 0.5 0.5 0.25 123 3 1.6: 2:4 2.56 0.7 630 3A 1.6: 2: 4 2.56 0.36 630 4 1.6: 4: 4 2.56 0.7 630 5 1: 2.3: 4.5 4.5 0.36 1120 5A 1: 2.3: 4.5 4.5 0.31 320 System No. [A.sub.G1] [A.sub.G2] [Cd.sub.A1] [Cd.sub.A2] 1 126 154 0.64 0.79 2 77 102 0.63 0.83 3 450 573 0.72 0.91 3A 630 630 1 1 4 590 600 0.94 0.95 5 1086 1098 0.97 0.98 5A 320 320 1 1 System No. [A.sub.g] tot [Cd.sub.A-ave] 1 280 0.71 2 179 0.73 3 1023 0.81 3A 1260 1 4 1190 0.94 5 2184 0.98 5A 2240 1 T: gate thickness, W: gate width, [A.sub.Gt]: theoretical gate area (one gate, [mm.sup.2]), [A.sub.G] tot: average of calculated area ([mm.sup.2]), [Cd.sub.ave]: average of areal coefficient of discharge Table 4: Flow Rate From Gates and Total Coefficient of Discharge System No Sprue:Runner:Gate G/S [(T/W).sub.G] [Q.sub.t] [Q.sub.1] 1 1: 0.9: 0.8 0.8 - 1117 230.6 2 1: 0.7: 0.5 0.5 - 1117 152.5 3 1.6: 2: 4 2.56 0.7 1117 265.5 3A 1.6: 2: 4 2.56 0.36 1117 365.5 4 1.6: 4: 4 2.56 0.7 1117 359.9 5 1: 2.3:4.5 4.5 0.36 1117 488.7 5A 1: 2.3: 4.5 4.5 0.31 1117 144 System No [Q.sub.22 [Q.sub.tot] [Cd.sub.1] [Cd.sub.2] Cd 1 255.6 486.2 0.41 0.46 0.43 2 173.4 325.9 0.27 0.31 0.29 3 372.5 638 0.48 0.67 0.57 3A 378 743.4 0.66 0.68 0.67 4 414 774 0.64 0.74 0.69 5 527 1015.7 0.87 0.94 0.91 5A 147.2 896 0.9 0.92 0.8 * * average of coefficient of discharge for 7 gates [Q.sub.tot]: total flow rates from gates ([mm.sup.3]/s x [10.sup.3]), [Q.sub.t]: theoretical flow rate form all the gate ([mm.sup.3]/s x [10.sup.3]), Cd: total coefficient of discharge
Table 2 presents the gate velocity and velocity coefficient of discharge. The measured gate velocities were lower than the theoretical values. This was due to variations in flow direction and friction between the liquid and the walls.
The conventional coefficient of discharge (CdV), calculated via the ratio of measured to theoretical velocity, is influenced by the following parameters:
Gate Distance to the Sprue--In the pressurized aluminum gating systems, the velocity and CdV is greater for the first gate(s) than for the others. Considering the flow patterns, two factors should be considered:
The fluid is bounced back upstream after complete filling of the runner, resulting in turbulence in front of the second gate. This results in lower fluid velocity at the second gate.
Vena contracta due to the back pressure is less in the second gate, which results in lower velocity.
In non-pressurized systems, the velocity through the second gate is slightly greater than the first gate. This might be associated with an inappropriate tapering of the runner. For controlling velocity, stepping may be better than straight tapering.
Effects of Gating Ratios--By increasing the gate-to-sprue area ratio (G/S), or runner-to-sprue area ratio (R/S) for the systems with similar G/S values, fluid velocity decreases. Accordingly, CdV increases. This is predictable, as fluid is less turbulent through increasing cross-sectional areas. This trend is more apparent in pressurized systems where turbulence and friction are greater.
Effect of Area to Perimeter of Gate--By increasing the G/S ratio, the area/perimeter ratio (A/P) of the gate also is increased. In systems with similar G/S ratios (3 vs. 3A and 5 vs. 5A), A/P is another important parameter. For instance, Systems 3 and 3A have similar G/S and R/S values and similar velocities. However, the area/perimeter ratio of the gate is lower for 3A, which means the system has a greater gate periphery and consequently higher friction and reduced CdV (Fig. 3).
[FIGURE 3 OMITTED]
The ratio of filled area over the total cross-sectional area of the gate is called areal coefficient of discharge and is less than or equal to one. The areal coefficient of discharge is shown in Table 3. As indicated, all the gates except those in Systems 3A and 5A were not completely filled, which could be attributed to vena contracta in pressurized systems and vortex formation in non-pressurized systems. Parameters affecting the areal coefficient of discharge are:
Gate Distance from Sprite--In each system, the areal coefficient of discharge is greater for the second gate than the first (except Systems 3A and 5A, with CdA = 1), due to back pressure from filling the runner.
Effects of the Gating Ratios (G/S and R/S) and Gate Velocities--Generally, by increasing the G/S ratio, gate velocity decreases. As a result, the area filled by the fluid increases. This translates into a higher CdA (Fig. 4). However, pressurized systems have lower filled areas because of vena contracta. This effect reduces the filled area more significantly than vortex formation in non-pressurized systems.
[FIGURE 4 OMITTED]
Effect of Thickness-to-Width Ratio of the Gate--In non-pressurized systems of equal G/S values, reducing the gate thickness/width ratios will result in greater filling of the gate area and consequently an increased CdA (Fig. 5).
[FIGURE 5 OMITTED]
The overall coefficient of discharge is a function of filled area and gate velocity and is calculated by multiplying the velocity by the areal coefficients of discharge (Cd= CdVxCdA). The closer the Cd value is to 1, the lower the turbulence will be in the system and more predictable mold filling will be.
In an attempt to rationalize the total coefficient of discharge, two points should be emphasized. In actual aluminum casting processes, the optimized flow conditions should give rise to the maximum flow rate at a minimum velocity. As shown in Table 6, non-pressurized systems have higher flow rates. Non- pressurized systems also have closer flow rate values to theoretical flow rates, and systems designed based on the critical gate velocity could represent enhanced gating systems.
About the Author
Shahrooz Nafisi is a research engineer for Evraz Inc. NA, Regina, Saskatchewan, Canada.
RELATED ARTICLE: Rules of the Flow
Remember the following rules of thumb when developing your next aluminum gating system.
1. In pressurized systems, fluid flow is turbulent and almost no time is available for tranquilization. Most of the contaminated metal and inclusions enter the mold cavity. Vena contracta (effective choke) in junctions also leads to a smaller area and higher velocity. Non-pressurized systems are less turbulent and have a lower gate velocity. By locating the gates above the runners, the chance of contaminated liquid and inclusions entering in the mold is minimized. Vortexes often are observed at the gates, which is associated with the velocity and gate ratio (width-to-length ratio). This phenomenon can be eliminated by reducing the aforementioned ratio and/or decreasing the velocity.
2. In pressurized systems, the velocity coefficient of discharge for the first gate is more than the second. The inverse is true for non-pressurized systems. For non-pressurized systems, this small difference can be due to improper tapering.
3. By increasing the gate-to-sprue area (G/S) and runner-to-sprue area (R/S) ratios, the velocity coefficient of discharge (CdV) in all systems is increased. In the case of similar values for G/S, another effective parameter is the ratio of A/P. The higher the A/P value, the higher the CdV. In the case of identical values for (A/P)G, the (A/P) R is a better substitute.
4. The overall coefficient of discharge (Cd) is calculated from experimental data and is defined as the ratio of the actual flow rate to the theoretical flow rate by multiplying the velocity coefficient of discharge by the areal coefficient of discharge. Closer values to 1 provide less turbulence and more predictable mold filling. With increased G/S values and velocity reductions, Cd becomes closer to one. Considering the similar values for G/S, another helpful parameter is the thickness-to-width ratio of the gate.
Visit www.moderncasting.com for video of the water modeling experiments described in this article.
Shahrooz Nafisi, Evraz Inc. NA, Regina, Saskatchewan, Canada
|Printer friendly Cite/link Email Feedback|
|Date:||Sep 1, 2010|
|Previous Article:||Cast Connex sources Cast construction part: Cast Connex Corp. worked with its metal casting supplier to produce a construction frame connector that...|
|Next Article:||Climate, Copenhagen and energy.|