I have used surface-dried sand mold casting as the main production route for high-pressure steel castings because it gives me a practical balance between mold strength, surface quality, dimensional control, and production economy. In my experience, a surface-dried sand mold is not simply a green sand mold with a little more moisture removed, nor is it a fully dry sand mold that consumes large amounts of fuel and electric power. It is a controlled intermediate process. The mold surface is dried to a certain depth, while the bulk of the mold remains relatively simple and economical to prepare. This approach has allowed me to produce dense, pressure-tight components that must pass strict hydrostatic testing and must remain free from leakage, sweating, and pressure drop.
When I compare surface-dried sand mold casting with green sand molding, the differences are clear in production. A green sand mold has relatively high moisture content throughout the molding sand. That moisture can become steam when liquid steel enters the cavity, and the result is often blowholes, pinholes, sand burn-on, scabbing, and mold erosion. A surface-dried sand mold, by contrast, has a stronger surface layer with much lower moisture near the cavity face. Because the surface layer is stronger and drier, the liquid metal is less likely to react with moisture or to break down the mold surface. The casting therefore has a lower risk of gas defects, sticky sand, scabs, and erosion defects.
When I compare surface-dried sand mold casting with dry sand molding, the advantage is mainly economic and operational. A fully dry sand mold requires a long baking cycle, a large amount of fuel, and significant electric power for handling and ventilation. It also lengthens the production cycle and can make the working environment hotter and more difficult. Surface-dried sand mold casting saves fuel and electric power, shortens the production cycle, and improves the working environment and labor conditions. For single-piece and small-batch production, this is a major advantage. For high-pressure castings, however, the process must be controlled much more carefully than for ordinary commercial castings.
I have applied surface-dried sand mold casting to many industrial and mining replacement parts. Later, I received a larger batch of high-pressure castings used in oil-field wellhead equipment. These castings are safety-related and pressure-related components. The main family included three types of parts. The most representative part was a large four-way fitting. I will use that part as the example to explain how I selected process parameters, how I met the drawing requirements, and how I produced pressure-tight, dense castings through sand mold casting.
Technical requirements and process implications. The material was a cast steel grade suitable for pressure service. The net weight was significant, and the technical requirements were strict. First, the casting had to be annealed to relieve casting stress. Second, after rough machining, the blank had to be quenched and tempered to achieve the required hardness. Third, the gasket ring groove was not allowed to contain any small sand hole, blowhole, or other defect that could affect sealing performance. Fourth, the part had to pass a hydrostatic test for a specified duration with no leakage, no sweating, and no pressure drop. These requirements mean that the foundry cannot treat the part as an ordinary sand casting. Every step of sand mold casting must be designed to reduce shrinkage porosity, gas porosity, inclusions, and mold-related surface defects.
| Technical Requirement | Main Risk if Not Controlled | Process Response in Sand Mold Casting |
|---|---|---|
| Stress relief annealing | Residual stress, dimensional movement, cracking during machining | Control cooling after pouring, use proper annealing cycle, avoid sharp section transitions |
| Quenching and tempering after rough machining | Hardness variation, insufficient toughness | Consistent chemistry, proper heat treatment, clean casting surface, adequate stock allowance |
| No defects in gasket ring groove | Leakage, poor sealing, scrap after machining | Place ring groove in a dense region, use chills and risers, avoid core-related defects |
| Hydrostatic test without leakage or sweating | Through-wall porosity, shrinkage porosity, cracks, gas holes | Directional solidification, sufficient riser feeding, low-moisture surface-dried mold, controlled pouring |
| Pressure drop limitation | Micro-porosity and interconnected voids | Use high feeding efficiency, clean steel, proper gating, chills, and sufficient riser contact |
Mold selection and general comparison. I selected surface-dried sand mold casting because it provides a higher surface strength and a lower surface-layer moisture content than green sand molding. The surface layer resists erosion and scabbing better. It also reduces the chance that steam from the mold will enter the steel and form blowholes. Compared with dry sand molding, the surface-dried mold saves fuel and electric power, shortens the production cycle, and improves labor conditions. The following table summarizes the comparison I use when selecting the process for high-pressure castings.
| Mold Type | Surface Strength | Surface Moisture | Gas Defect Risk | Fuel and Power Use | Production Cycle | Suitability for High-Pressure Castings |
|---|---|---|---|---|---|---|
| Green sand mold | Low to moderate | High throughout mold | High | Low | Short | Limited unless gates and risers are very conservative |
| Surface-dried sand mold | High at cavity surface | Low at cavity surface | Moderate to low | Moderate | Moderate | Good for single-piece and small-batch pressure castings |
| Dry sand mold | High throughout | Very low | Low | High | Long | Excellent but more expensive |
For the large four-way fitting, the geometry required a horizontal symmetric parting. I placed the large and small flanges on the side of the mold so that all four through holes lay on the parting plane. This arrangement gave me two important benefits. First, it made core support easier because the cores could be positioned and seated more reliably. Second, it improved venting because gases could escape along the parting plane and through the core prints. In sand mold casting, core support and venting are not minor details. If a core shifts or if gas cannot escape, the result can be a distorted hole, a gas pocket, or a penetration defect that later becomes a leak path under pressure.
Pattern shrinkage and dimensional control. I used a pattern shrinkage rule that accounted for external dimensions and internal dimensions separately. The external shrinkage and internal shrinkage are not always identical because the mold and core restraints differ. The external pattern dimension can be written as:
$$L_p = L_c \left(1 + S_e\right)$$
where L_p is the pattern dimension, L_c is the final casting dimension at room temperature, and S_e is the external shrinkage allowance. For internal dimensions, I used:
$$L_{p,i} = L_{c,i} \left(1 + S_i\right)$$
where S_i is the internal shrinkage allowance. In practice, I allowed the external shrinkage and internal shrinkage according to the cast steel grade, the section size, and the restraint from cores. The following table gives the general allowances I used for this class of sand mold casting.
| Dimension Type | Allowance Symbol | Typical Range | Reason for Difference |
|---|---|---|---|
| External length and width | S_e | 1.5% to 2.0% | Free contraction of outer surfaces is relatively less restrained |
| Internal holes and cores | S_i | 1.0% to 1.8% | Core restraint and cooling conditions differ from outer surfaces |
| Flange thickness | S_t | 1.2% to 2.0% | Section thickness affects cooling rate and contraction |
| Overall height | S_h | 1.5% to 2.2% | Height is affected by mold ramming and bottom restraint |
I also considered machining allowance separately. The gasket ring groove and sealing surfaces required enough stock to clean up without exposing subsurface porosity. If the machining allowance is too small, a minor surface defect can become a leak path after machining. If the allowance is too large, the machining cost increases and the dense skin of the casting may be removed. For pressure castings, I prefer a moderate but consistent allowance and then rely on riser and chill design to keep the subsurface region dense.
Riser design for sand mold casting. The riser is the most important feeding feature in sand mold casting for high-pressure components. I used a waist-cone open riser, placed on top of the two large flanges. The riser must remain liquid longer than the casting section it feeds. The basic modulus concept is:
$$M = \frac{V}{A}$$
where M is the modulus, V is the volume, and A is the heat-transfer surface area. For a sound casting, the riser modulus M_r must be greater than the casting modulus M_c at the feeding location:
$$M_r = k M_c$$
where k is a safety factor. For cast steel, k is often in the range of 1.2 to 1.5, depending on the feeding distance and the required soundness level. Solidification time follows Chvorinov’s rule:
$$t_s = B M^2$$
where t_s is the solidification time, B is the mold constant, and M is the modulus. Therefore, the ratio of riser solidification time to casting solidification time is:
$$\frac{t_{s,r}}{t_{s,c}} = \left(\frac{M_r}{M_c}\right)^2$$
This equation explains why I must not simply make the riser visually large. I must make the riser modulus large enough and keep the riser hot enough to feed the last liquid shrinkage into the casting. For the large four-way fitting, I calculated the thermal node circle at the junction between the flange and the conical body. The average wall thickness and the height of the section determined the local modulus. I then used a proportional method to estimate the riser width.
Let the increased thickness be represented by a variable, and let the average wall thickness be t. The thermal node diameter at the flange-to-cone junction is D_n. The riser width W_r can be estimated from a ratio that depends on the casting height H and the node diameter. In general form, I used:
$$W_r = c_1 D_n + c_2 t$$
where c_1 and c_2 are empirical coefficients. When the ratio of casting height to node diameter falls within a certain range, the riser width can be selected from a design chart. I chose the final riser width so that the riser would freeze after the flange. The chord length of the riser was then calculated from the flange circumference and the riser width. The riser taper was set so that the riser could be withdrawn easily and so that the hot top would remain open to the atmosphere.
| Riser Parameter | Symbol | Basis of Selection | Purpose |
|---|---|---|---|
| Riser type | Waist-cone open riser | High feeding efficiency, easy top pouring | Keep riser hot and open to air |
| Riser location | Top of large flanges | Hottest feeding points | Feed the flange and conical body junction |
| Riser modulus | M_r | M_r = k M_c | Ensure later solidification than casting |
| Riser width | W_r | Chart and proportional method | Provide sufficient liquid volume |
| Riser chord length | L_c | Flange circumference and W_r | Match flange geometry |
| Riser taper | alpha | Pattern practice | Easy withdrawal and feeding |
The casting had two main annular hot nodes, so I placed two open risers. The riser weight, casting weight, and pouring weight determine the process yield. I used the following relation:
$$\eta = \frac{W_c}{W_c + W_r + W_g} \times 100\%$$
where eta is the process yield, W_c is the casting weight, W_r is the riser weight, and W_g is the gating system weight. A high yield is desirable, but for pressure castings I do not sacrifice feeding safety just to improve yield. The riser must be large enough to compensate for liquid shrinkage and to promote directional solidification. In my trials, the two risers and the gating system added significant weight, but the resulting castings passed the hydrostatic test consistently. That confirmed the feeding design was more important than a small gain in yield.
Directional solidification and chills. To ensure that the casting solidifies toward the risers, I placed an external chill on the lower outer side of each large flange. The chill thickness and outer dimensions were selected according to the local modulus. The purpose of the chill is to increase the cooling rate in the flange region so that the flange solidifies before the riser loses its feeding ability. The chill also helps reduce shrinkage porosity in the flange, which is close to the sealing surface.
$$Q = \rho c_p V \Delta T$$
In this equation, Q is the heat absorbed by the chill, rho is density, c_p is specific heat, V is the chill volume, and Delta T is the temperature rise of the chill. The chill must have enough heat capacity to extract heat from the casting section without becoming saturated too quickly. I used a chill thickness that was sufficient for the local section and I made sure the chill surface was flat, clean, and free from rust. If the chill surface is dirty or rusty, the result can be gas defects, cold shuts, or incomplete fusion at the chill face. In sand mold casting, an external chill is only effective if it makes intimate contact with the mold cavity and if the mold coating does not insulate it excessively.
| Chill Location | Chill Type | Key Control | Expected Effect |
|---|---|---|---|
| Lower outer side of large flanges | External chill | Flat, clean, rust-free surface | Accelerate flange solidification |
| Below riser contact | External chill | Correct thickness and position | Promote directional solidification toward riser |
| Sealing surface region | External chill | Avoid chill gaps and moisture | Reduce shrinkage porosity near ring groove |
I also controlled the thermal gradients by adjusting the gating system and the pouring sequence. The goal was not to chill the entire casting uniformly, because that would make the riser ineffective. The goal was to create a temperature gradient from the far flange regions toward the risers. In sand mold casting, this gradient is the key to pressure tightness. If the last liquid is isolated in a region far from the riser, shrinkage porosity will form there. If that region is later machined into a sealing surface, the casting will fail the hydrostatic test.
Gating system design. I placed the gating system in the lower flask at the flange plate. To avoid direct impingement of the steel stream on the cores, I directed the internal gates downward and tangentially. I used an open gating system. The cross-sectional area ratio was selected as:
$$A_s : A_r : A_g = 1 : 2 : 3$$
where A_s is the sprue area, A_r is the runner area, and A_g is the total gate area. This open ratio reduces the velocity of the metal entering the mold, which helps prevent mold erosion and core displacement. I also used a sprue brick for the down sprue. The choke area was calculated from the pouring time and the metallostatic head. The basic equation for choke area is:
$$A_c = \frac{W}{\rho C_d t_p \sqrt{2 g H_m}}$$
where A_c is the choke area, W is the weight of liquid metal, rho is density, C_d is the discharge coefficient, t_p is pouring time, g is gravitational acceleration, and H_m is the average metallostatic head. I controlled the pouring time to avoid both too-fast filling and too-slow filling. If the fill is too fast, the mold can erode and the core can move. If the fill is too slow, the metal front can freeze before the mold is full.
| Gating Component | Design Choice | Reason |
|---|---|---|
| Sprue | Sprue brick, calculated choke | Control filling rate and avoid turbulence |
| Runner | Open ratio, 1:2:3 | Reduce velocity and distribute metal evenly |
| Ingate | Downward inclined, tangential | Prevent direct core impingement and promote smooth filling |
| Location | Lower flask flange plate | Provide favorable thermal gradient and easy assembly |
I also considered the Reynolds number to keep the flow as smooth as practical. The Reynolds number can be estimated by:
$$Re = \frac{\rho v D}{\mu}$$
where v is velocity, D is hydraulic diameter, and mu is dynamic viscosity. In sand mold casting, fully laminar flow is difficult to achieve, but reducing turbulence is important. Turbulent flow can entrain gas, erode sand, and create inclusions. The tangential ingates and the open gating ratio helped reduce these problems. I inspected the gating system after each trial to confirm that the gates did not cut the mold surface and that the cores remained in position.
Sand formulation and coating. The molding sand for surface-dried sand mold casting must have a relatively coarse grain size. A coarse sand provides good permeability, which allows gas to escape from the mold surface during pouring. At the same time, the surface layer must be rammed tightly and uniformly so that the mold cavity is strong and resistant to metal erosion. I used a facing sand and a coating system. The facing sand contained silica sand, clay, and other additives. The coating was applied in two passes. After the first coating, I waited about one hour before applying the second coating. Then I allowed the mold to dry naturally for several hours.
| Material | Ingredient | Typical Function | Control Point |
|---|---|---|---|
| Facing sand | Silica sand | Refractory base | Coarse grain size for permeability |
| Facing sand | Clay or bentonite | Binder for green strength | Maintain sufficient surface strength |
| Facing sand | Water | Plasticity during ramming | Low moisture after surface drying |
| Coating | Quartz powder | Refractory surface layer | Uniform coverage, no runs |
| Coating | Binder | Adhesion to sand surface | Do not over-dilute |
| Coating | Carrier liquid | Application consistency | Allow proper drying between coats |
The physical properties of the molding sand are just as important as the recipe. I measured moisture, green compressive strength, permeability, and surface stability. The following table shows the general property targets I used for sand mold casting of pressure castings.
| Property | Symbol | Target Range | Why It Matters |
|---|---|---|---|
| Moisture | w | Low after surface drying | Reduce steam-related gas defects |
| Green compressive strength | sigma_g | Sufficient for handling and ramming | Prevent mold wall movement and erosion |
| Permeability | P | Moderate to high | Allow gas escape during pouring |
| Surface hardness | H_s | Uniform and high at cavity face | Resist scabbing and burn-on |
| Dried layer thickness | x_d | About 10 mm or as required | Provide dry, strong surface layer |
After ramming, I made sure the mold cavity surface was tight and uniform. Loose spots can cause sand wash, scabs, and rough surfaces. After repairing the mold, I applied the first coating. The second coating followed after about one hour. Then I allowed the mold to dry naturally for approximately four hours before baking. The natural drying step removes part of the carrier liquid and reduces the risk of surface cracking during torch baking.
Mold baking and assembly. After natural drying, I used a diesel torch to bake the mold. The first baking lasted about twenty minutes. After an interval of about ten minutes, I placed the cores and performed a second baking for about fifteen minutes. After another interval of about ten minutes, I checked the dried layer thickness. The target was a dried layer of about ten millimeters. Once the dried layer reached the requirement, I cleaned the mold cavity of loose sand, ash, and debris. Then I closed the mold and prepared for pouring.
| Step | Operation | Time or Interval | Control Requirement |
|---|---|---|---|
| 1 | Natural drying after coating | About 4 hours | No wet patches, no coating runs |
| 2 | First torch baking | About 20 minutes | Uniform surface heating |
| 3 | Interval before core setting | About 10 minutes | Avoid thermal shock to cores |
| 4 | Second torch baking | About 15 minutes | Dry the surface after core placement |
| 5 | Interval before inspection | About 10 minutes | Allow temperature equalization |
| 6 | Check dried layer | As required | Target about 10 mm |
| 7 | Cleaning and closing | Before pouring | Remove loose sand and ash |
After closing, I required the mold to be poured within four hours. If the mold remains open or closed for too long, moisture can migrate back into the surface layer. This is especially dangerous for sand mold casting of high-pressure steel parts because moisture can cause blowholes and pinholes. If a longer delay is unavoidable, I reopen the mold and bake it again. I never allow a surface-dried mold to sit overnight without protection and re-drying. The surface-dried layer is only effective if it remains dry until the steel enters the cavity.
Pouring practice. The pouring temperature for surface-dried sand mold casting should not be too high. If the temperature is too high, the liquid steel absorbs moisture from the mold surface and generates gas defects. High temperature also increases the risk of sand burn-on, which makes cleaning difficult and can damage the casting surface. I prefer low-temperature fast pouring. The exact pouring temperature depends on the steel grade and the casting geometry. In general, I set it only as high as necessary to fill the mold and feed the risers. A useful relation is:
$$T_p = T_L + \Delta T$$
where T_p is the pouring temperature, T_L is the liquidus temperature, and Delta T is the superheat. For many cast steels, the superheat is kept moderate. I controlled the pouring time within a narrow window. The pouring time can be approximated by:
$$t_p = K \sqrt{W}$$
where K is an empirical coefficient and W is the casting weight. This relation helps me scale the pouring time from one casting to another. For the large four-way fitting, I used a pouring time that allowed complete filling without excessive turbulence. The exact value was established through trials and then fixed in the process sheet.
| Pouring Stage | Action | Control Point | Reason |
|---|---|---|---|
| Initial fill | Pour through gating system | Steady stream, no interruption | Prevent cold shut and entrapment |
| Riser approach | Stop when riser reaches a set height | About 80% to 90% of riser height | Prepare for hot topping |
| Sprue plugging | Plug sprue with sand | Prevent backflow and air entry | Maintain riser pressure |
| Top pouring | Pour steel from riser top | Fill riser completely | Increase riser temperature and feeding |
| Hot topping | Add charcoal powder and straw ash | Cover riser surface | Insulate and keep riser open to air |
| Re-pouring | Supplement riser after an interval | About 5 minutes | Compensate for shrinkage |
| Final covering | Cover with ash again | Do not let skin form early | Improve feeding efficiency |
I used a specific riser-feeding sequence. When the steel reached about eighty to ninety percent of the riser height, I stopped pouring through the gating system. I plugged the sprue with sand. Then I poured additional steel directly into the riser from the top. Just before the riser was full, I added charcoal powder and straw ash for insulation. After about five minutes, I supplemented the riser once more. After the supplement, I covered the riser again with ash. This practice keeps the riser steel liquid longer and maintains a connection between the riser and the atmosphere. It increases the feeding efficiency of the riser and reduces shrinkage porosity in the casting. In sand mold casting of pressure castings, this simple practice often makes the difference between a passing and a failing hydrostatic test.
Cooling, cleaning, and heat treatment. After pouring, I allowed the casting to cool in the mold under controlled conditions. I did not shake out too early, because early shakeout can cause cracking, especially in complex flanges and cored holes. After shakeout, I removed the risers and gates, cleaned the surface, and inspected the casting. The castings then received stress relief annealing. After rough machining, they were quenched and tempered to the required hardness. The gasket ring groove was machined after heat treatment and inspected carefully. Any indication of porosity in the sealing region was grounds for rejection.
| Post-Pouring Step | Purpose | Key Control |
|---|---|---|
| In-mold cooling | Prevent cracking and allow complete solidification | Avoid early shakeout |
| Shakeout and cleaning | Remove sand, risers, and gates | Do not damage casting surface |
| Stress relief annealing | Relieve residual stress | Follow specified heating and cooling cycle |
| Rough machining | Prepare for quenching and tempering | Maintain sufficient stock for final dimensions |
| Quenching and tempering | Achieve hardness and toughness | Control temperature, time, and cooling rate |
| Final machining of ring groove | Ensure sealing surface quality | Inspect for porosity and inclusions |
| Hydrostatic test | Verify pressure tightness | No leakage, sweating, or pressure drop |
Defect prevention in sand mold casting. The defects I watch most closely in pressure castings are blowholes, pinholes, sand burn-on, scabs, sand wash, shrinkage porosity, and core-related defects. Each defect has a different cause and a different countermeasure. The following table summarizes the troubleshooting logic I use for surface-dried sand mold casting.
| Defect | Likely Cause | Countermeasure |
|---|---|---|
| Blowholes | High moisture in mold surface, poor venting, turbulent filling | Dry surface layer, improve core vents, reduce pouring turbulence |
| Pinholes | Steel gas content, mold reaction, excessive superheat | Clean steel, lower pouring temperature, dry mold |
| Sand burn-on | High pouring temperature, reactive sand, insufficient coating | Lower temperature, improve coating, use appropriate refractory sand |
| Scabs | Weak surface layer, high moisture, rapid heating | Increase surface ramming, dry surface layer, uniform baking |
| Sand wash | High metal velocity, weak mold surface, poor gating | Open gating ratio, tangential ingates, stronger surface |
| Shrinkage porosity | Insufficient riser feeding, wrong thermal gradient, missing chills | Increase riser modulus, add chills, improve directional solidification |
| Core shift | Weak core support, high pouring force, poor core assembly | Improve core prints, symmetric parting, secure core setting |
| Gas holes near core | Core binder gas, poor core venting, moisture in core | Bake cores, improve venting, reduce core moisture |
Case study: large four-way fitting. The large four-way fitting is a high-pressure oil-field wellhead component with two large flanges and two smaller flanges. The four through holes lie on the parting plane. I used a horizontal symmetric parting. The large flanges are on the side of the mold. The risers are waist-cone open risers placed on top of the two large flanges. External chills are placed on the lower outer side of each large flange. The gating system enters from the lower flask at the flange plate, with downward inclined tangential ingates. The molding sand is a coarse silica sand with a facing sand and a two-coat refractory coating. The mold surface is dried naturally and then baked with a diesel torch to a dried layer of about ten millimeters.
| Process Feature | Selection for Large Four-Way | Function |
|---|---|---|
| Molding method | Surface-dried sand mold casting | High surface strength, low moisture, reduced gas defects |
| Parting plane | Horizontal symmetric parting | Core support and venting |
| Flange position | Large and small flanges on side | All four through holes on parting plane |
| Riser type | Waist-cone open riser | High feeding efficiency and easy top pouring |
| Riser location | Top of two large flanges | Feed annular hot nodes |
| Chill | External chill on lower outer flange | Promote directional solidification |
| Gating | Open system, downward tangential ingates | Reduce core impingement and turbulence |
| Pouring | Low-temperature fast pour | Reduce gas defects and burn-on |
| Riser feeding | Top pouring with ash insulation and re-pouring | Increase riser efficiency |
| Hydrostatic test | Required for every casting | Verify pressure tightness |
For this part, I used the modulus method to check the riser. The thermal node at the flange-to-cone junction was the controlling point. I calculated the local modulus and selected a riser modulus with a safety factor. I then checked the solidification time ratio:
$$\frac{t_{s,r}}{t_{s,c}} = \left(\frac{M_r}{M_c}\right)^2 > 1$$
This condition must be satisfied at every feeding path. If the ratio is less than one, the riser may freeze before the casting section, and shrinkage porosity will form. I also checked the feeding distance from the riser to the furthest point of the flange. For a sand mold casting with chills, the effective feeding distance is longer than without chills, but it is not unlimited. The chill must be placed so that the last freezing region is connected to the riser through a liquid channel. If the chill is too strong or too far from the riser, it can isolate a section and create a internal defect.
Hydrostatic testing and acceptance. The final acceptance test is the hydrostatic test. The casting is filled with water, pressurized to the specified pressure, and held for the specified time. The test requires no leakage, no sweating, and no pressure drop. Sweating is often the first sign of micro-porosity. A casting may look perfect on the outside but still show sweating under pressure. This is why I pay so much attention to the gasket ring groove and the flange section. These are the regions where micro-porosity is most likely to be exposed by machining. If the sand mold casting process is not controlled, the ring groove may contain small sand holes or gas holes that are invisible before machining but become leak paths under pressure.
| Test Parameter | Requirement | Process Implication |
|---|---|---|
| Test medium | Water | Reveals through-wall and micro-porosity |
| Pressure | Specified by drawing | Must be supported by dense casting section |
| Hold time | Specified duration | Allows sweating and pressure drop to appear |
| Leakage | None allowed | Requires sound flange and ring groove |
| Sweating | None allowed | Requires low micro-porosity and no interconnected voids |
| Pressure drop | None allowed | Requires global pressure tightness |
I have produced thousands of large four-way fittings by this surface-dried sand mold casting route. They have passed the hydrostatic test and met the drawing requirements. The process has proven successful. The key is not to rely on one single measure, but to combine correct riser sizing, proper gating, external chills, clean and dry mold surfaces, controlled pouring, and hot-top feeding of the risers. For pressure castings, a larger riser alone is not a guarantee. In fact, blindly increasing riser size and pads can waste metal and can even worsen the thermal gradient if the feeding path is not designed correctly.
Process rules I follow for sand mold casting of pressure castings. First, I calculate the modulus at the hot nodes and design the riser to freeze after the casting. Second, I use chills to establish a favorable thermal gradient toward the riser. Third, I design the gating system to fill the mold smoothly and avoid core impingement. Fourth, I use a surface-dried sand mold with a coarse, permeable sand and a strong, dry surface layer. Fifth, I control coating thickness and baking to achieve a dry layer of the required depth. Sixth, I pour at a relatively low temperature and a controlled fast rate. Seventh, I feed the riser from the top with insulation and re-pouring. Eighth, I test every pressure casting hydrostatically. These rules are simple, but they must be followed consistently.
| Rule | Action | Effect on Sand Mold Casting |
|---|---|---|
| 1 | Calculate modulus and riser size | Ensure riser feeds the last liquid |
| 2 | Use chills at hot spots | Promote directional solidification |
| 3 | Design smooth gating | Reduce turbulence and core damage |
| 4 | Control sand composition | Improve permeability and surface strength |
| 5 | Dry and bake the mold surface | Reduce moisture-related gas defects |
| 6 | Pour at controlled temperature and time | Reduce burn-on, blowholes, and cold shuts |
| 7 | Hot-top and re-pour the riser | Increase feeding efficiency |
| 8 | Hydrostatic test every casting | Verify pressure tightness |
Mathematical summary of feeding and solidification. For a sound sand mold casting, the following relationships summarize the design logic. The modulus is:
$$M = \frac{V}{A}$$
The solidification time is:
$$t_s = B M^2$$
The riser-to-casting modulus ratio is:
$$M_r = k M_c$$
The solidification time ratio is:
$$\frac{t_{s,r}}{t_{s,c}} = \left(\frac{M_r}{M_c}\right)^2$$
The process yield is:
$$\eta = \frac{W_c}{W_c + W_r + W_g} \times 100\%$$
The choke area is:
$$A_c = \frac{W}{\rho C_d t_p \sqrt{2 g H_m}}$$
The pouring time relation is:
$$t_p = K \sqrt{W}$$
The drying depth can be approximated by:
$$x_d = K_d \sqrt{t_d}$$
where K_d is a drying coefficient and t_d is drying time. These equations are not a substitute for experience, but they provide a consistent framework for sand mold casting process design. When I combine them with practical trials, I can produce high-pressure castings that meet strict acceptance standards.
Conclusion. Surface-dried sand mold casting is an effective process for high-pressure steel castings when it is properly controlled. It provides a higher surface strength and lower surface moisture than green sand molding, which reduces blowholes, burn-on, scabs, and erosion. It saves fuel and electric power compared with dry sand molding, shortens the production cycle, and improves working conditions. For single-piece and small-batch production of pressure castings, it is a practical and reliable route. The large four-way fitting is a good example. By using symmetric horizontal parting, core support on the parting plane, waist-cone open risers, external chills, open gating with tangential ingates, a coarse sand with a dried surface layer, low-temperature fast pouring, and hot-top riser feeding, I have produced thousands of pressure-tight castings that pass hydrostatic testing. The experience confirms that sand mold casting can be used for demanding pressure service when the process parameters are calculated, controlled, and verified.