In my work on the second stage moving blade, I treated the component as a high-value precision casting in which geometry, internal core stability, shell integrity, and melting practice are tightly coupled. The blade is large, thin in local sections, and geometrically complex. Its envelope is approximately 351 mm × 168 mm × 89 mm, and the airfoil length is about 203 mm. Because the tip region is far from the gating feed path and because the internal cavity contains a fine grid, the casting route must be controlled with unusual care. I use the term precision casting throughout this article because every decision—from wax injection to shell drying to vacuum pouring—must serve dimensional accuracy, internal soundness, and repeatability.
1. Casting Geometry and Quality Challenges
I began by mapping the quality risks that define this precision casting. The blade is hollow, and its internal cavity is formed by a ceramic core. The exhaust trailing slot has a specified width of 1.2 mm ± 0.2 mm. That slot is not merely a dimensional feature; it also influences water-flow testing, because the flow area controls the pressure drop and the effective discharge coefficient. A tip slot that is too narrow can lower the flow rate, while an oversized slot can increase it. Therefore, the core, shell, and metal all must hold their positions within a narrow process window.
The tip region is especially critical. The distance between the tip slot and the airfoil surface is about 2.6 mm. This creates a narrow shell cavity during investment casting. If the shell slurry or stucco bridges across that gap, the shell becomes locally thick and rigid, or it may fail to fill completely. Both outcomes harm the precision casting: a bridged shell can reduce heat transfer and cause incomplete filling, while a weak shell can crack during pouring. The tenon pocket on the back side is also deep, and its geometry makes drying difficult. If moisture remains trapped there, the shell can delaminate, dry unevenly, or generate excess metal that is difficult to remove.
| Feature | Nominal Value | Process Risk | Control Focus |
|---|---|---|---|
| Overall envelope | 351 mm × 168 mm × 89 mm | Dimensional distortion | Wax cooling, shell strength, pouring temperature |
| Airfoil length | 203 mm | Tip deformation during solidification | Gating, shell preheat, thermal gradient |
| Tip slot to airfoil distance | About 2.6 mm | Shell bridging, weak tip shell | Fine stucco, slurry viscosity, sand removal |
| Exhaust trailing slot | 1.2 mm ± 0.2 mm | Flow-area deviation | Core position, wax injection, dimensional scan |
| Tenon pocket | Deep pocket | Poor drying, shell delamination, excess metal | Compressed-air drying, local sand cleaning |
| Internal cavity | Fine grid structure | Core cracking, core shift | Core support, wax filling, low-stress handling |
For a precision casting of this size, the thermal field is not uniform. I approximate the local cooling behavior with the one-dimensional heat conduction relation:
$$ \frac{\partial T}{\partial t} = \alpha \frac{\partial^2 T}{\partial x^2} $$
where T is temperature, t is time, x is distance, and α is thermal diffusivity. The diffusivity itself is defined by:
$$ \alpha = \frac{k}{\rho c_p} $$
Here, k is thermal conductivity, ρ is density, and cp is specific heat capacity. In my experience, the tip region behaves as a thin, extended section with a high surface-to-volume ratio. It loses heat quickly, but the shell and core can also impose local thermal resistance. That competition is why I do not treat the tip as a simple chill; I treat it as a coupled precision casting problem involving shell thickness, core mass, and gating feed.
2. Wax Pattern and Ceramic Core Strategy
The ceramic core is the internal skeleton of this precision casting. Its grid structure is complex on the exhaust side, and the cylindrical connection near the tenon-side core print is fragile. The thinnest part of the tip region is about 3 mm. During wax injection, the core experiences pressure, shear, and thermal contraction from the surrounding wax. If the wax shrinks onto the core while the core is unsupported, the core can crack. Even a hairline crack may later become a metal penetration path or a dimensional deviation.
I therefore filled the weak positions with R839 runner wax before injection. The purpose was not to change the final internal cavity but to provide temporary mechanical support during wax pressing. The filling operation had to be complete. A virtual connection—wax that appears to touch the core but does not actually support it—is unacceptable because it gives a false sense of security and may allow the core to flex. In my trials, I inspected the filled zones visually and by handling before placing the core into the die.
| Core Condition | Risk During Wax Injection | My Countermeasure | Inspection Method |
|---|---|---|---|
| Thin tip wall about 3 mm | Cracking under wax pressure | Support with R839 runner wax | Visual check and manual stability check |
| Complex exhaust-side grid | Local core deflection | Balanced injection and support filling | Post-wax core extraction after wax removal |
| Tenon-side cylindrical core print | Fracture at connection | Fill and stabilize the transition | Inspection before and after pressing |
| Long hollow body | Core shift along the axis | Controlled flow rate and die temperature | Section check and dimensional scan |
I adjusted the wax injection parameters as follows: wax cylinder temperature 80 °C ± 5 °C, nozzle temperature 80 °C ± 5 °C, injection pressure 5 bar ± 2 bar, and flow rate 40 cc/s. Before using the ceramic core, I pressed wax without the core to evaluate the surface quality of the wax pattern. With this flow rate, the wax surface showed only slight flow lines when no cold slug and no core were used. When I inserted both the core and the cold slug, the flow lines disappeared, but local incomplete filling appeared near the exhaust side.
To understand that incomplete filling, I performed short-shot trials at 5 s, 10 s, 12 s, 13 s, 14 s, and 15 s. The filling behavior showed that during the first 10 s, the basin side and back side filled at nearly the same speed. In the final stage, the back side encapsulated faster than the basin side. Because the basin-side exhaust edge has a very small space, the wax could not enter that last region, and the basin-side exhaust edge remained unfilled. This is a classic precision casting problem: the last-to-fill region is not necessarily the farthest region; it is the region where flow resistance and solidification resistance meet.
| Short-Shot Time | Observed Filling State | Interpretation |
|---|---|---|
| 5 s | Early fill, major channels open | Flow front still developing |
| 10 s | Basin and back sides comparable | Balanced initial filling |
| 12 s | Back side begins to lead | Local resistance increases on basin side |
| 13 s | Basin-side exhaust edge still open | Last-to-fill region identified |
| 14 s | Partial basin-side exhaust edge | Wax mobility becomes limiting |
| 15 s | Incomplete basin-side exhaust edge | Solidification and narrow gap dominate |
The flow rate is related to volume and time by:
$$ Q = \frac{V}{t} $$
where Q is volumetric flow rate, V is volume, and t is filling time. For a given shot, the local velocity in a channel can be approximated by:
$$ v = \frac{Q}{A} $$
where A is the cross-sectional area. Near the exhaust edge, A becomes very small. If Q is held constant, v must rise, which increases shear rate and pressure loss. For a narrow rectangular gap, the pressure drop is strongly sensitive to gap height. I use the following proportional relation to guide my reasoning:
$$ \Delta P \propto \frac{\mu L Q}{h^3 w} $$
Here, ΔP is pressure loss, μ is dynamic viscosity, L is flow length, h is gap height, and w is gap width. This relation explains why a small reduction in local gap height can cause a large increase in required pressure. It also explains why raising wax temperature helps: higher temperature lowers μ, so the wax can continue to move in the final stage. However, I must not raise the wax temperature excessively, because higher temperature increases shrinkage during cooling and can impose additional compressive stress on the ceramic core.
My solution was to increase the wax cylinder temperature within a controlled range. The goal was to preserve enough fluidity at the end of filling without overheating the wax. I also maintained the die and nozzle temperatures within the stated window. After optimizing the parameters, I pressed ten blades. After complete cooling, I removed the wax by chemical or thermal wax removal and inspected the ceramic cores. No core cracking was found. This result was important because it confirmed that the combination of local wax support and controlled injection had protected the fragile core in this precision casting.
| Wax Injection Parameter | Target Value | Reason for Control |
|---|---|---|
| Wax cylinder temperature | 80 °C ± 5 °C | Balance fluidity and shrinkage |
| Nozzle temperature | 80 °C ± 5 °C | Prevent premature freezing at gate |
| Injection pressure | 5 bar ± 2 bar | Fill narrow sections without core damage |
| Flow rate | 40 cc/s | Control shear and filling sequence |
| Core support wax | R839 runner wax | Stabilize thin core sections |
| Verification batch | 10 blades | Confirm core integrity after wax removal |
3. Gating and Cluster Design
For the second stage moving blade, the length in the Z direction is about 351 mm. I selected a top pouring arrangement because it can provide adequate feeding for this height. The riser design was based on the casting envelope and the tenon cross section. The tenon section length in the Z direction is about 80 mm. I set the riser width to about 75 mm and the other principal width to about 40 mm. These dimensions were not arbitrary; I adjusted them according to the metallurgical quality required at the tenon and according to the feed path available through the blade root.
I used a split riser approach. First, I assembled the wax pattern with the riser connected to the tenon. Then I joined that subassembly to the remaining cluster using bonding wax. After assembly, the transition between the part and the riser was made smooth. A smooth transition reduces turbulence during pouring and helps avoid localized hot spots or cold shuts. In precision casting, the gating system is not only a delivery channel; it is also a thermal design that controls solidification order.
| Gating Feature | Design Choice | Function |
|---|---|---|
| Pouring position | Top pouring | Feed the long Z-direction height |
| Riser width | About 75 mm | Match tenon envelope and feed demand |
| Secondary riser width | About 40 mm | Adjust local thermal mass |
| Assembly method | Split riser with bonding wax | Simplify handling and improve transition |
| Transition | Smooth fillet | Reduce turbulence and hot spots |
To assess feeding, I use the Niyama criterion as a qualitative indicator:
$$ Niyama = \frac{G}{\sqrt{R}} $$
where G is the temperature gradient and R is the cooling rate. A higher Niyama value generally indicates a lower risk of shrinkage porosity. The top pouring design increases the thermal gradient along the blade, while the riser supplies liquid metal to the last solidifying regions. I also consider the solidification range of the alloy, because a wide freezing range increases the risk of dispersed porosity.
The local solid fraction can be approximated by:
$$ f_s = 1 – \left( \frac{T_L – T}{T_L – T_S} \right)^{1/(1-k)} $$
where fs is solid fraction, TL is liquidus temperature, TS is solidus temperature, T is local temperature, and k is the partition coefficient. This relation reminds me that feeding must remain active until the local solid fraction is high enough to block liquid flow. The riser size and position therefore must be matched to the solidification sequence of the blade, not merely to the external envelope.
4. Shell-Making Process
The shell is the mold that gives the precision casting its final shape. For this blade, two shell problems dominated my trials: poor drying at the deep tenon pocket and bridging at the narrow tip space. The tenon pocket on the back side is deep, and during the transition layer drying stage, moisture can remain trapped. I therefore directed compressed air to that region after the shell had dried for about 2 h. The air flow helped remove moisture and equalize drying. Before dipping, I also cleaned loose sand from the basin-side and back-side tenon pockets with compressed air. If loose sand remains, the next slurry layer may not bond properly, and the shell can delaminate during pouring.
The tip region is even more delicate. The narrow space is about 2.6 mm. If the stucco grain is too large, it can bridge across the gap. A bridge creates a thick, rigid shell bridge that changes local heat transfer and can also cause a weak spot at the tip. In severe cases, the bridge can block the shell from forming a continuous cavity. I therefore selected a finer stucco for the early layers. For the second and third layers, I used 30–60 mesh mullite sand. Before each layer, I cleaned loose sand from the narrow region and then continued shell building. This step is simple but essential for precision casting quality.
| Shell Layer | Slurry Powder and Specification | Slurry Viscosity | Stucco Material and Specification |
|---|---|---|---|
| 1 | Zircon sand, 325 mesh | 36 s | Zircon sand, 100 mesh |
| 2–3 | Mullite powder, 200 mesh | 15 s | Mullite sand, 30–60 mesh |
| 4–10 | Mullite powder, 200 mesh | 12 s | Mullite sand, 16–30 mesh |
| 11 | Mullite powder, 200 mesh | 10 s | Sealing slurry |
The shell was built to ten layers and then sealed. This layering scheme balances strength, permeability, and surface finish. The first layer provides a fine surface against the wax pattern. The second and third layers control the narrow tip space and reduce bridging. The fourth through tenth layers build mechanical strength. The final sealing layer closes the surface and reduces dusting or loose particles. For a precision casting of this size, the shell must resist the thermal shock of pouring while still allowing enough heat transfer to promote directional solidification.
Drying can be approximated by a diffusion-like relation:
$$ \frac{\partial C}{\partial t} = D \frac{\partial^2 C}{\partial x^2} $$
where C is moisture concentration, D is effective diffusivity, and x is depth. The characteristic drying time scales with the square of the diffusion distance:
$$ t_d \sim \frac{L^2}{D} $$
For the deep tenon pocket, L is large, so td increases rapidly. That is why I used compressed air locally. The air reduces the boundary-layer resistance and helps remove moisture from the pocket. I also avoid excessive slurry accumulation in the pocket, because a thick slurry layer increases L and slows drying further.
For the narrow tip gap, the bridging risk can be related to the ratio of stucco particle size to gap width:
$$ B = \frac{d_p}{h} $$
where B is a bridging index, dp is characteristic particle diameter, and h is gap height. When B is too large, particles can lodge across the gap and form a bridge. I keep B low by using finer stucco in the early layers and by cleaning the gap before each dip. This is a practical expression of precision casting control: the shell must fit the geometry, not the geometry fit the shell.
| Shell Problem | Cause | My Process Adjustment | Expected Result |
|---|---|---|---|
| Deep tenon pocket drying | Long diffusion path and trapped moisture | Compressed air after 2 h drying; clean loose sand before dipping | Uniform drying and reduced delamination |
| Tip gap bridging | Stucco particle size close to gap width | Use 30–60 mesh mullite sand in layers 2–3; clean gap before each layer | Reduced bridging and stronger tip shell |
| Shell delamination | Loose sand and poor interlayer bonding | Compressed-air cleaning of basin and back tenon pockets | Improved layer adhesion |
| Insufficient shell strength | Low layer count or poor sealing | Ten layers plus sealing | Withstand pouring and handling |
5. Melting and Pouring Practice
The blade was made from K444 alloy, a nickel-based precipitation-hardening equiaxed cast superalloy. It is used below about 900 °C and offers good high-temperature strength and hot-corrosion resistance. Its melting range is about 1 262–1 319 °C. Based on that range and on the blade geometry, I set the pouring temperature to 1 440–1 470 °C. The vacuum level was controlled within 10 Pa. The refining temperature was 1 520 °C ± 10 °C, and the refining time was at least 3 min. The pouring speed was controlled within 2 s, and after pouring the mold was held for at least 10 min before being taken out for cooling. The casting was cooled for more than 6 h before shell removal.
I preheated the shell with the pouring cup facing downward. The preheat temperature was 1 100 °C ± 10 °C, and the holding time was at least 1 h. This preheat reduces thermal shock, improves fillability, and helps control the thermal gradient at the tip. If the shell is too cold, the metal may freeze before the narrow tip and exhaust slot are filled. If the shell is too hot, the solidification front may become too diffuse, and shrinkage porosity may increase. The preheat therefore must be matched to the alloy, the shell thickness, and the gating design.
| Melting or Pouring Parameter | Target Value | Process Purpose |
|---|---|---|
| Alloy | K444 nickel-based cast superalloy | High-temperature strength and corrosion resistance |
| Melting range | About 1 262–1 319 °C | Define superheat window |
| Shell preheat temperature | 1 100 °C ± 10 °C | Reduce thermal shock and improve filling |
| Shell preheat holding time | At least 1 h | Uniform mold temperature |
| Refining temperature | 1 520 °C ± 10 °C | Homogenize melt and remove inclusions |
| Refining time | At least 3 min | Allow reactions and floatation |
| Pouring temperature | 1 440–1 470 °C | Balance fluidity and solidification |
| Vacuum level | Within 10 Pa | Prevent oxidation and gas defects |
| Pouring speed | Within 2 s | Fill mold before premature freezing |
| Post-pour holding | At least 10 min | Complete solidification under controlled conditions |
| Cooling before shell removal | More than 6 h | Avoid cracking and warping |
The superheat is defined as:
$$ \Delta T_{sh} = T_p – T_L $$
where Tp is pouring temperature and TL is liquidus temperature. For K444, using TL ≈ 1 319 °C and Tp = 1 440–1 470 °C, the superheat is approximately:
$$ \Delta T_{sh} \approx 121 \text{ to } 151 \text{ °C} $$
This superheat range is enough to fill the thin exhaust slot and the tip region, but not so high that it causes excessive grain growth or core reaction. I also consider the Reynolds number in the gating channel to estimate flow regime:
$$ Re = \frac{\rho v D}{\mu} $$
where ρ is density, v is velocity, D is hydraulic diameter, and μ is dynamic viscosity. In vacuum pouring, the actual flow is transient and strongly affected by the mold geometry, but the Reynolds number helps me judge whether the gating design may produce excessive turbulence. I prefer a smooth, progressive filling front because it reduces the risk of oxide entrapment and cold shuts in this precision casting.
Heat transfer from the metal to the shell can be represented by Newton’s law of cooling:
$$ q = h A (T_m – T_s) $$
where q is heat flux, h is heat-transfer coefficient, A is area, Tm is metal temperature, and Ts is shell temperature. The tip region has a high surface-to-volume ratio, so A is relatively large. That accelerates local cooling and can cause the tip to freeze before the last feed metal arrives. The preheated shell and the top pouring design help offset this effect.
6. Inspection and Results
After pouring, cleaning, and cutting the riser, I inspected the castings. No underfill or missing metal was found. After shell removal, I used blue-light scanning to evaluate the surface dimensions. The dimensional requirement was +0.75 mm/−0.50 mm. The measured results satisfied the tolerance. This confirmed that the shell layer count and shell strength were sufficient to hold the blade shape during pouring and cooling. It also confirmed that the wax injection parameters and core support strategy had maintained the internal cavity position within an acceptable range.
After dimensional inspection, I repaired surface defects and used wire cutting to form the tenon teeth. Then I performed fluorescent penetrant inspection. No unacceptable indications were found, and the surface condition was good. I also performed radiographic inspection according to the relevant standard. No unacceptable indications were found. In my assessment, the combination of process controls produced a precision casting that met the required quality level.
| Inspection Stage | Method | Acceptance Focus | Result |
|---|---|---|---|
| Dimensional survey | Blue-light scanning | Profile tolerance +0.75 mm/−0.50 mm | Satisfied |
| Surface inspection | Fluorescent penetrant inspection | No unacceptable surface indications | No unacceptable defects |
| Internal inspection | Radiographic inspection | No unacceptable internal defects | No unacceptable defects |
| Visual inspection | Post-cleaning visual check | No underfill, no missing metal | Satisfied |
| Flow-related feature | Dimensional check of exhaust slot | 1.2 mm ± 0.2 mm | Controlled by core and wax process |
For the water-flow requirement, the exhaust slot and internal grid must remain stable. The flow rate through a slot can be approximated by:
$$ Q = C_d A \sqrt{\frac{2 \Delta P}{\rho}} $$
where Cd is discharge coefficient, A is flow area, ΔP is pressure difference, and ρ is fluid density. Because A depends on the slot height and width, the 1.2 mm ± 0.2 mm tolerance directly affects flow capacity. A small deviation in slot height changes A and therefore changes Q. This is why I treat the core position and the wax injection filling pattern as critical variables in this precision casting.
The dimensional deviation of the casting can be summarized as:
$$ \Delta_{total} = \Delta_{wax} + \Delta_{core} + \Delta_{shell} + \Delta_{shrink} + \Delta_{thermal} $$
where Δwax is wax pattern deviation, Δcore is ceramic core deviation, Δshell is shell deformation, Δshrink is solidification shrinkage, and Δthermal is thermal contraction. To meet the final tolerance, I must control each term. The wax parameters control Δwax. The core support and handling control Δcore. The shell layers, drying, and preheat control Δshell. The gating and pouring temperature control Δshrink. The cooling practice controls Δthermal.
| Deviation Source | Main Process Lever | Observed Control Strategy |
|---|---|---|
| Wax pattern | Temperature, pressure, flow rate | 80 °C ± 5 °C, 5 bar ± 2 bar, 40 cc/s |
| Ceramic core | Support filling and handling | R839 wax support at weak sections |
| Shell deformation | Layer count, stucco size, drying | Ten layers plus sealing; fine stucco at tip |
| Solidification shrinkage | Riser design and pouring temperature | Top pouring, split riser, 1 440–1 470 °C |
| Thermal contraction | Cooling time and mold preheat | Preheat 1 100 °C ± 10 °C; cool over 6 h |
7. Discussion of Process Interactions
In my view, the most important lesson from this work is that the process variables are not independent. A change in wax temperature affects wax shrinkage, which affects core stress, which affects core cracking, which affects the final wall thickness and flow area. A change in shell stucco size affects bridging, which affects shell strength, which affects dimensional stability, which affects the final precision casting tolerance. A change in pouring temperature affects fillability, which affects the exhaust slot, which affects water flow. Therefore, I optimize the process as a system rather than as a set of isolated steps.
I can express the coupling between wax shrinkage and core stress in a simplified form. The thermal strain of the wax is:
$$ \varepsilon_{wax} = \alpha_{wax} \Delta T_{wax} $$
The stress imposed on the core is approximately:
$$ \sigma_{core} \approx E_{wax} \varepsilon_{wax} $$
where Ewax is the effective elastic modulus of the solidifying wax. If the wax temperature is too high, ΔTwax increases, and the stress on the core rises. If the wax temperature is too low, the wax may freeze before the narrow exhaust edge is filled. The solution is a window, not a single point. In my trials, the window was centered around 80 °C ± 5 °C with a flow rate of 40 cc/s and a pressure of 5 bar ± 2 bar.
The shell drying process is similarly coupled. The moisture diffusion time is:
$$ t_d \sim \frac{L^2}{D} $$
If the tenon pocket has a depth L that is several times larger than the nominal wall thickness, the drying time can increase by an order of magnitude. Compressed air reduces the external mass-transfer resistance and effectively increases D at the surface. Cleaning loose sand before dipping reduces the local thickness and prevents delamination. These actions are simple, but they are essential for a reliable precision casting.
The tip bridging problem can be controlled by the bridging index:
$$ B = \frac{d_p}{h} $$
When B approaches or exceeds a critical value, bridging becomes likely. By using 30–60 mesh mullite sand in the second and third layers, I reduced dp and kept B low. I also cleaned the gap before each layer. This prevented the stucco from accumulating in the narrow space and forming a bridge. The result was a stronger tip shell and a more reliable precision casting.
8. Final Process Window
The final process window that I established for this blade is summarized below. It is the result of wax trials, shell trials, pouring trials, and inspection feedback. I use this window as the baseline for production and for further improvement.
| Process Area | Parameter | Final Set Point or Range |
|---|---|---|
| Wax injection | Wax cylinder temperature | 80 °C ± 5 °C |
| Wax injection | Nozzle temperature | 80 °C ± 5 °C |
| Wax injection | Injection pressure | 5 bar ± 2 bar |
| Wax injection | Flow rate | 40 cc/s |
| Core preparation | Weak-section support | R839 runner wax, fully connected |
| Cluster assembly | Gating type | Top pouring with split riser |
| Cluster assembly | Riser width | About 75 mm and about 40 mm |
| Shell making | Layer 1 | Zircon sand 325 mesh, viscosity 36 s, stucco zircon 100 mesh |
| Shell making | Layers 2–3 | Mullite powder 200 mesh, viscosity 15 s, mullite sand 30–60 mesh |
| Shell making | Layers 4–10 | Mullite powder 200 mesh, viscosity 12 s, mullite sand 16–30 mesh |
| Shell making | Layer 11 | Mullite powder 200 mesh, viscosity 10 s, sealing |
| Shell drying | Tenon pocket | Compressed air after 2 h; clean loose sand before dipping |
| Shell drying | Tip narrow space | Clean before each layer; fine stucco in early layers |
| Shell preheat | Temperature and time | 1 100 °C ± 10 °C for at least 1 h, cup down |
| Melting | Refining temperature | 1 520 °C ± 10 °C |
| Melting | Refining time | At least 3 min |
| Pouring | Pouring temperature | 1 440–1 470 °C |
| Pouring | Vacuum level | Within 10 Pa |
| Pouring | Pouring speed | Within 2 s |
| Post-pour | Holding time | At least 10 min |
| Cooling | Before shell removal | More than 6 h |
9. Conclusions
I conclude that the second stage moving blade can be produced as a qualified precision casting by combining controlled wax injection, local ceramic core support, a well-balanced gating design, a fine-stucco shell strategy, and a carefully bounded melting and pouring practice. The key wax injection parameters are wax cylinder temperature 80 °C ± 5 °C, nozzle temperature 80 °C ± 5 °C, pressure 5 bar ± 2 bar, and flow rate 40 cc/s. The shell is built to ten layers and then sealed, with 30–60 mesh mullite sand in the second and third layers to reduce tip bridging. The melting practice uses a refining temperature of 1 520 °C ± 10 °C, a pouring temperature of 1 440–1 470 °C, a refining time of at least 3 min, a pouring speed within 2 s, and a post-pour holding time of at least 10 min.
With these controls, the castings showed no underfill or missing metal after cleaning and riser removal. Blue-light scanning confirmed that the profile dimensions met the +0.75 mm/−0.50 mm requirement. Fluorescent penetrant inspection and radiographic inspection found no unacceptable indications. The exhaust slot, internal grid, and tip region remained within the process window required for water-flow testing. In my assessment, the process is robust enough for further production and provides a solid foundation for continued improvement of this precision casting.