In my development work on post-processing systems for stereolithography apparatus (SLA) additive manufacturing, I have repeatedly observed that the cleaning stage is one of the most critical and least standardized steps in the production chain for precision casting resin molds. When a resin mold is printed for precision casting, uncured resin remains on both external and internal surfaces because the liquid photopolymer has a finite viscosity and cannot fully drain before post-curing. If this residual liquid is not removed in a controlled manner, the precision casting mold can suffer from dimensional drift, surface defects, incomplete burnout, ash residues, and reduced casting quality. My objective in this work was therefore to develop a high-performance equipment platform that removes residual liquid from precision casting resin molds efficiently, safely, and reproducibly.
The core problem can be expressed in simple mass and surface terms. For a precision casting resin mold, the amount of residual resin after printing is a function of surface area, drainage time, viscosity, and local geometry. I define the residual mass as:
$$R = \frac{M_r}{A_s}$$
where \(R\) is the residual mass per unit surface area, \(M_r\) is the mass of remaining resin, and \(A_s\) is the effective surface area of the precision casting mold. The cleaning efficiency of any process can then be expressed as:
$$\eta_c = \frac{M_0 – M_r}{M_0} \times 100\%$$
Here, \(M_0\) is the initial mass of residual liquid before cleaning, and \(M_r\) is the mass after cleaning. For precision casting, the target is not merely high \(\eta_c\); the process must also preserve mold hardness, dimensional accuracy, and surface integrity. This dual requirement distinguishes precision casting resin mold cleaning from general 3D-printed part washing.
Process Context and Engineering Challenge
SLA additive manufacturing has expanded from prototype models into precision casting, ceramic processing, and complex mold production. For precision casting, the resin mold must later be heated or burned out, so the wall thickness is often kept near 3 mm to reduce material consumption and ash generation. Thin walls are beneficial for casting, but they are vulnerable to solvent attack, mechanical brushing, and prolonged immersion. I found that the traditional open soaking method creates a difficult trade-off: sufficient soaking removes resin, but excessive soaking softens the mold and compromises the precision casting process.
The conventional cleaning workflow typically includes a solvent such as high-concentration industrial alcohol, an open cleaning tank, an exhaust fan, and manual brushes. In my analysis, this workflow has several structural weaknesses. First, it requires a large volume of alcohol, often 20 kg or more for a moderate precision casting mold. Second, the alcohol is stored and used in an open or semi-open environment, which increases fire and explosion risk. Third, the cleaning time is difficult to control for different geometries, so thin precision casting resin molds may become soft. Fourth, manual brushing generates debris, damages surfaces, and creates inconsistent results. Fifth, the exhaust system is often insufficient and does not guarantee a safe working environment.
| Cleaning Aspect | Traditional Soaking Method | Open Manual Method | Impact on Precision Casting |
|---|---|---|---|
| Solvent volume | 20–22 kg or more | Variable, often excessive | High storage risk and high cost |
| Cleaning time | 40–50 minutes | Operator dependent | Thin walls may soften |
| Surface quality | Acceptable but inconsistent | Brush marks and debris possible | May affect precision casting surface finish |
| Dimensional stability | Risk of swelling or deformation | Localized mechanical damage | Can exceed precision casting tolerance |
| Safety | Low due to open alcohol | Low due to vapor exposure | Unacceptable for production environments |
| Floor space | Large | Moderate to large | Limits cell layout for precision casting |
I therefore set a clear design direction: replace open immersion with a closed, spray-based, PLC-controlled cleaning platform. The equipment must integrate solvent delivery, recovery, filtration, ventilation, sensing, and human-machine interaction into one unit. It must be able to clean precision casting resin molds with thin walls while avoiding internal cavity flooding and avoiding long solvent contact. It must also reduce the amount of alcohol stored and used per cycle.
Design Requirements and Functional Positioning
I organized the development around four functional pillars: integration, high efficiency, high safety, and ease of operation. These pillars are not independent. For example, a closed chamber improves safety, but it also requires automated ventilation and solvent recovery to maintain efficiency. A spray process reduces solvent volume, but it requires precise control of nozzle timing, pressure, and drainage. A PLC and HMI make the process repeatable, but they also require reliable sensors and fail-safe logic.
| Design Pillar | Requirement | Implementation Direction | Precision Casting Benefit |
|---|---|---|---|
| Integration | Combine all cleaning steps on one platform | Closed chamber, solvent unit, control cabinet, HMI | Consistent cleaning for precision casting molds |
| High efficiency | Reduce cycle time and solvent use | Repeated spray and blow drying | Faster precision casting mold turnaround |
| High safety | Minimize alcohol vapor and liquid exposure | Negative pressure, explosion-proof electrical design | Safer production of precision casting resin molds |
| Ease of operation | Reduce operator skill requirements | Graphical HMI, one-touch spray cycles | Repeatable precision casting quality |
I also defined quantitative targets. The cleaning equipment should use less than 3 kg of alcohol per cleaning cycle for common precision casting molds, complete a cycle within 10–20 minutes, leave no visible surface residue, avoid measurable wall softening, and keep the working area free of alcohol vapor accumulation. The equipment should occupy about 3 m² and should be compatible with multiple precision casting mold sizes.
Existing Cleaning Process Limitations
Before designing the new platform, I analyzed the existing process in detail. The cleaning agent is typically 95% industrial alcohol by volume because it dissolves uncured resin and evaporates without leaving a film. However, alcohol has a low flash point, high volatility, and a wide explosive range when mixed with air. Large open tanks intensify these hazards because the liquid surface is continuously exposed. The exhaust fan may remove some vapor, but it does not prevent local concentration gradients or operator exposure.
I also examined the cleaning container. A simple metal tank or ultrasonic cleaner may hold enough alcohol to immerse a 500 mm × 500 mm × 500 mm precision casting mold, but this requires approximately 100 kg of alcohol. Such a volume is unacceptable for a safe production environment. In addition, the tank does not provide continuous filtration, so dissolved resin accumulates, cleaning effectiveness drops, and the solvent must be replaced frequently. The brushes used for manual cleaning shed bristles and can scratch delicate precision casting mold surfaces.
| Existing Element | Problem Observed | Consequence for Precision Casting |
|---|---|---|
| Open alcohol tank | Large exposed liquid surface | Fire risk, vapor exposure, high solvent consumption |
| Immersion cleaning | Uncontrolled contact time | Thin walls soften, dimensions change |
| Manual brushing | Non-uniform force and brush debris | Surface defects on precision casting molds |
| Simple exhaust | Incomplete vapor capture | Unsafe room concentration, health risk |
| No filtration loop | Resin accumulation in solvent | Poor repeatability for precision casting |
From this analysis, I concluded that the cleaning process must be transformed from an open, batch-soaking operation into a closed, spray-and-blow operation with controlled solvent dosing. The precision casting mold should never be submerged. Instead, the residual resin should be dissolved and carried away by a sequence of directed sprays, followed by air blow and vapor extraction.
System Architecture of the New Cleaning Equipment
I designed the equipment as an integrated platform with four main modules: negative-pressure ventilation and filtration, cleaning treatment, interactive control, and solvent control. The overall appearance and composition are shown in the figure below.

The negative-pressure ventilation module uses a variable-frequency fan, temperature and humidity sensors, and flow guide plates. The fan speed adapts to the measured alcohol concentration, maintaining a slight negative pressure in the working zone so that vapor does not escape into the room. The extracted gas passes through a filter that removes alcohol before clean air is discharged.
The cleaning treatment module provides a closed cleaning chamber and switchable cleaning tools. Multiple nozzles are installed at the top of the chamber, and the spray strategy can be configured according to the precision casting mold geometry. For hard-to-reach areas, an alcohol spray gun and a compressed-air gun can be used to ensure that internal and external surfaces are clean and dry.
The interactive control module includes an HMI touch screen, buttons, and a foot switch on the outside, plus electrical components and a PLC on the inside. Both sections use explosion-proof design principles to improve operational safety. The solvent control module supplies, recovers, and filters the alcohol. It includes a solvent reservoir, multi-position level sensors, gas supply lines, a pressure regulator, a weighing sensor, a circulation pump, and an alcohol concentration sensor.
| Module | Main Components | Primary Function | Role in Precision Casting Cleaning |
|---|---|---|---|
| Negative-pressure ventilation | Variable-frequency fan, concentration sensor, filter | Adaptive exhaust and vapor removal | Keeps precision casting mold cleaning safe |
| Cleaning treatment | Nozzles, spray gun, air gun, closed chamber | Spray and blow cleaning | Removes residual resin without immersion |
| Interactive control | HMI, PLC, buttons, foot switch | Parameter setting and automated control | Makes precision casting cleaning repeatable |
| Solvent control | Reservoir, level sensors, pump, weighing sensor | Quantitative supply, recovery, filtration | Reduces alcohol use for precision casting molds |
Hardware Implementation and Component Selection
For the control core, I selected a Siemens PLC Smart 200 as the data analysis and processing controller, and an IE700 touch screen as the human-machine interface. The PLC program is written in ladder logic, while the HMI is configured with WinCC. This combination allows all core parameters to be displayed on a graphical interface, monitored automatically, and used to trigger actions. I paid particular attention to explosion protection because alcohol vapor may be present inside the cleaning chamber and solvent cabinet.
The solvent reservoir has a capacity of 30 L. This is sufficient for at least 10 cleaning cycles, which is about three times the efficiency of the previous open cleaning method. The reservoir is pressurized by compressed air, and the pressure regulator can be set between 0.5 and 0.8 MPa. This pressure drives the spray nozzles. Three position sensors monitor the high, middle, and low liquid levels. The high level indicates that filling should stop, the middle level indicates remaining usable volume, and the low level indicates that refilling is required.
The waste liquid collector is mounted on a weighing sensor. The mass data is sent to the PLC, which compares it with a preset maximum waste limit, typically 90% of the actual collector capacity. When the limit is approached, the PLC triggers an alarm and prevents further cleaning until the waste is removed. This prevents overflow and keeps the precision casting cleaning process under control.
| Component | Specification or Range | Signal Type | Control Function |
|---|---|---|---|
| PLC | Siemens Smart 200 | Digital and analog I/O | Central logic and safety interlocks |
| HMI | IE700 touch screen | Ethernet or serial | Parameter entry and visualization |
| Alcohol concentration sensor | 0–100% lower explosive limit | 0–10 V DC analog | Fan speed control |
| Fan encoder | Pulse output | Digital high-speed counter | Closed-loop fan speed feedback |
| Level sensors | High, middle, low | Digital discrete | Solvent filling and refill alarm |
| Weighing sensor | Waste collector mass | Analog or digital | Waste liquid limit control |
| Pressure regulator | 0.5–0.8 MPa | Pneumatic | Spray pressure regulation |
Negative-Pressure Adaptive Ventilation
The ventilation subsystem is essential for safety and for precision casting quality. If alcohol vapor remains in the chamber, it can condense on the mold surface and leave streaks. If it escapes into the room, it creates a health and explosion hazard. I therefore implemented an adaptive negative-pressure strategy. The alcohol concentration sensor sends a 0–10 V DC analog signal to the PLC. The PLC converts this signal into an alcohol concentration value from 0 to 100% of the lower explosive limit. The fan speed is then adjusted according to the concentration.
The fan encoder returns a pulse signal to the PLC, allowing closed-loop speed regulation. I used a PID control law of the form:
$$u(t) = K_p e(t) + K_i \int_0^t e(\tau) d\tau + K_d \frac{de(t)}{dt}$$
Here, \(e(t)\) is the difference between the desired concentration and the measured concentration, and \(u(t)\) is the fan speed command. The proportional term responds to the current error, the integral term removes steady-state offset, and the derivative term reduces overshoot. In practice, I tuned the controller so that the chamber remains at a slight negative pressure without excessive air exchange, which would otherwise increase alcohol evaporation and energy consumption.
The pressure drop through the ventilation path can be approximated by:
$$\Delta P = \left( f \frac{L}{D} + \zeta \right) \frac{\rho v^2}{2}$$
where \(f\) is the friction factor, \(L\) is the duct length, \(D\) is the hydraulic diameter, \(\zeta\) is the local loss coefficient, \(\rho\) is the air density, and \(v\) is the air velocity. I used this relation to size the ducting and to select a fan that can maintain the required negative pressure even when the filter begins to load with resin and alcohol.
| Concentration Range | Fan Speed Strategy | Negative Pressure Target | Safety Action |
|---|---|---|---|
| 0–10% LEL | Low speed | Slight negative pressure | Normal operation |
| 10–25% LEL | Medium speed | Moderate negative pressure | Increase ventilation |
| 25–50% LEL | High speed | Strong negative pressure | Alarm and limit spray |
| Above 50% LEL | Maximum speed | Maximum negative pressure | Stop cleaning and isolate solvent |
Solvent Quantitative Supply, Recovery, and Filtration
The solvent control module is the heart of the high-performance cleaning equipment. Instead of immersing the precision casting mold, I use a repeated spray-and-blow sequence. The solvent is stored in a sealed reservoir and delivered by compressed air. The spray volume is controlled by valve timing and pressure. The mass balance for the solvent can be written as:
$$M_{t+1} = M_t + M_{in} – M_{spray} – M_{evap} – M_{recover}$$
where \(M_t\) is the solvent mass in the reservoir at time \(t\), \(M_{in}\) is the refill mass, \(M_{spray}\) is the mass delivered to the nozzles, \(M_{evap}\) is the mass lost to evaporation, and \(M_{recover}\) is the mass recovered from the chamber. The PLC tracks these quantities using level sensors, pressure data, and cycle counts. This allows the equipment to warn the operator when the solvent quality falls below the acceptable range.
The evaporation rate from the liquid surface and from wetted mold surfaces is approximated by:
$$\dot{m}_{evap} = k_e A_s (p_{sat} – p_{air})$$
where \(k_e\) is a mass transfer coefficient, \(A_s\) is the exposed surface area, \(p_{sat}\) is the saturation vapor pressure of alcohol, and \(p_{air}\) is the partial pressure in the surrounding air. By keeping the chamber closed and the ventilation adaptive, I reduce \(p_{air}\) near the mold and prevent uncontrolled vapor accumulation.
The circulation filter removes dissolved resin and particulate matter. The pressure drop across the filter is monitored. A rising pressure drop indicates filter loading. I use the following approximate relation for filter pressure drop:
$$\Delta P_f = \frac{\mu Q}{k A_f}$$
Here, \(\mu\) is the dynamic viscosity of the solvent, \(Q\) is the flow rate, \(k\) is the permeability of the filter medium, and \(A_f\) is the filtration area. When \(\Delta P_f\) exceeds a threshold, the HMI prompts maintenance. This ensures that the alcohol used for precision casting mold cleaning remains effective over multiple cycles.
| Solvent Parameter | Monitoring Method | Control Action | Precision Casting Benefit |
|---|---|---|---|
| High level | Position sensor | Stop filling | Prevents overfill and spillage |
| Middle level | Position sensor | Display remaining volume | Allows planned precision casting batches |
| Low level | Position sensor | Refill alarm | Avoids dry spraying |
| Waste mass | Weighing sensor | Stop at 90% capacity | Prevents overflow |
| Filter pressure drop | Differential pressure | Maintenance prompt | Maintains cleaning quality |
| Cycle count | PLC counter | Solvent replacement warning | Keeps precision casting repeatable |
Cleaning Chamber and Spray Strategy
The cleaning chamber is designed as a sealed enclosure with a transparent viewing window and an access door. The precision casting resin mold is placed on a rotating or fixed fixture, depending on geometry. The top-mounted nozzles can be grouped into zones. The HMI allows the operator to select a spray recipe for each zone, including spray time, delay time, pressure, and blow time. For complex precision casting molds, the operator can use a handheld alcohol spray gun and compressed-air gun through sealed glove ports or a controlled access opening.
The spray cleaning process is based on three mechanisms: dissolution, impingement, and drainage. Dissolution removes uncured resin by solvation. Impingement provides mechanical energy to detach viscous liquid from the surface. Drainage allows the dissolved resin to flow away before it redeposits. I model the spray flux as:
$$Q = \sum_{i=1}^{N} q_i \Delta t_i$$
where \(q_i\) is the volumetric flow rate of nozzle group \(i\), \(\Delta t_i\) is its activation time, and \(N\) is the number of groups. The impact pressure of a spray jet on the surface is approximately:
$$P_i = \frac{1}{2} \rho v^2$$
where \(\rho\) is the solvent density and \(v\) is the jet velocity. I selected nozzle types and pressures to achieve enough impact to remove resin without deforming thin precision casting mold walls. The dissolution rate can be represented as:
$$\frac{dM_r}{dt} = -k_d A_r (C_s – C)$$
where \(M_r\) is the residual resin mass, \(k_d\) is a dissolution coefficient, \(A_r\) is the wetted resin area, \(C_s\) is the saturation concentration of resin in alcohol, and \(C\) is the current concentration. By using fresh or filtered solvent and by replacing the liquid film frequently, I maintain a high \(C_s – C\) driving force.
| Spray Stage | Typical Time | Pressure Range | Purpose |
|---|---|---|---|
| Pre-wet | 10–20 s | 0.2–0.3 MPa | Soften surface resin |
| Main spray | 60–180 s | 0.4–0.6 MPa | Dissolve and detach residual resin |
| Rinse spray | 30–60 s | 0.3–0.5 MPa | Remove dissolved resin |
| Air blow | 60–120 s | 0.3–0.5 MPa | Dry surfaces and internal cavities |
| Vapor extraction | Continuous | Negative pressure | Prevent vapor escape |
PLC Control Logic and HMI Integration
I implemented the control logic so that the cleaning cycle is repeatable and safe. When the PLC is powered on, it first checks the solvent level and the waste collector mass. If the waste liquid mass is above the minimum threshold, the HMI prompts the operator to process the waste. The PLC then checks the solvent level. If the level is low, it enables the refill sequence. If the level is high, it prevents overfilling. During cleaning, the PLC starts and stops the circulation filter pump according to the solvent level, records the cycle count, and monitors the filter pressure drop.
The PLC scan can be represented in discrete form as:
$$y[k] = f(x[k], u[k])$$
where \(x[k]\) is the state vector, \(u[k]\) is the control input vector, and \(y[k]\) is the output vector at scan \(k\). The state vector includes liquid levels, waste mass, alcohol concentration, fan speed, valve states, and door status. The control inputs include spray valve commands, fan speed reference, pump command, and alarm outputs. The outputs include HMI messages, indicator lights, and safety interlocks.
The HMI provides a main operation interface with real-time values for alcohol concentration, reservoir level, waste mass, filter pressure, cycle count, and cleaning recipe. The operator can select a recipe, start a one-touch spray cycle, pause the cycle, or switch to manual mode. The foot switch allows hands-free confirmation during loading and unloading. All parameters are stored in the PLC and can be protected by access levels.
| HMI Screen | Displayed Data | Operator Action | Automatic Response |
|---|---|---|---|
| Main operation | Concentration, levels, waste mass, cycle count | Start, pause, reset | PLC executes cleaning recipe |
| Recipe setup | Spray time, pressure, blow time, nozzle zones | Edit and save | PLC uses new parameters |
| Maintenance | Filter pressure, pump hours, valve cycles | Reset counters | PLC schedules maintenance prompts |
| Alarm history | Time, code, description | Acknowledge | PLC logs event |
| Manual mode | Individual valve and fan controls | Jog or toggle | PLC enforces safety interlocks |
Mathematical Summary of Key Control Variables
To summarize the control and cleaning behavior, I use the following symbols and relations. These formulas guided the sizing of the reservoir, fan, filter, and spray nozzles for precision casting resin mold cleaning.
| Symbol | Meaning | Unit | Typical Range |
|---|---|---|---|
| \(M_r\) | Residual resin mass | kg | 0.01–0.5 |
| \(A_s\) | Effective surface area | m² | 0.1–2.0 |
| \(\eta_c\) | Cleaning efficiency | % | 95–100 |
| \(Q\) | Spray flux | L/min | 0.5–5.0 |
| \(P_i\) | Impact pressure | kPa | 50–500 |
| \(\Delta P\) | Ventilation pressure drop | Pa | 50–500 |
| \(C\) | Alcohol concentration | % LEL | 0–50 |
| \(M_{waste}\) | Waste liquid mass | kg | 0–20 |
The cleaning cycle time can be estimated as the sum of spray, dwell, blow, and extraction times:
$$t_{cycle} = \sum_{j=1}^{M} (t_{spray,j} + t_{dwell,j} + t_{blow,j}) + t_{extract}$$
For a typical precision casting resin mold, I found that \(t_{cycle}\) between 10 and 20 minutes is sufficient when the spray recipe is optimized. This is significantly shorter than the 40–50 minutes required by immersion cleaning. The throughput can be expressed as:
$$T_p = \frac{N}{t_{cycle}}$$
where \(N\) is the number of molds cleaned per batch. Because the chamber can hold one or more small precision casting molds, the effective throughput depends on fixture design and mold size.
Experimental Validation and Test Results
I validated the new cleaning method by comparing spray cleaning with traditional immersion cleaning on two precision casting resin molds: a small mold with a nominal diameter of 300 mm and a height of 350 mm, and a larger mold with greater internal complexity. The small mold was used for detailed measurement, while the larger mold was used to evaluate solvent consumption and safety. The results are summarized below.
| Test Item | Spray Cleaning | Immersion Cleaning | Improvement |
|---|---|---|---|
| Alcohol consumption | 2.0–2.3 kg | 20–22 kg | About 90% reduction |
| Cleaning time | 10–20 minutes | 40–50 minutes | About 60–75% reduction |
| Surface residue | No visible residue | No visible residue | Equivalent cleaning quality |
| Mold hardness | No measurable change | Wall surface softened | Better dimensional stability |
| Environmental impact | Closed chamber | Open working space | Lower vapor exposure |
| Floor space | About 3 m² | About 6 m² | 50% reduction |
| Operation complexity | Simple, HMI-guided | Complex, manual | Lower skill requirement |
| Safety | High | Low | Safer precision casting workflow |
I measured dimensional stability before and after cleaning using the relative change:
$$\delta = \frac{L_0 – L_f}{L_0} \times 100\%$$
For the spray-cleaned precision casting mold, \(\delta\) remained below 0.05%. For the immersion-cleaned mold, \(\delta\) exceeded 0.2% in thin-wall regions, indicating swelling and deformation. This difference is critical for precision casting because small dimensional errors can propagate into the final cast part.
I also measured surface hardness using an indentation method:
$$H = \frac{F}{A}$$
where \(F\) is the applied force and \(A\) is the indentation area. The spray-cleaned precision casting mold showed no significant hardness change, while the immersion-cleaned mold showed a measurable decrease in wall hardness. This confirms that avoiding prolonged solvent contact is essential for thin precision casting resin molds.
| Measurement | Spray-Cleaned Precision Casting Mold | Immersion-Cleaned Precision Casting Mold |
|---|---|---|
| Initial length \(L_0\) | 300.00 mm | 300.00 mm |
| Final length \(L_f\) | 299.94 mm | 299.40 mm |
| Relative change \(\delta\) | 0.02% | 0.20% |
| Hardness change | None detected | Visible softening |
| Surface residue | None | None |
| Internal cavity condition | Dry and clean | Liquid accumulation risk |
Safety Analysis and Risk Reduction
Safety is a primary design driver because alcohol is flammable, volatile, and explosive under certain conditions. I used a risk matrix to compare the traditional and new methods. The new equipment reduces risk through closed operation, quantitative solvent storage, adaptive ventilation, explosion-proof electrical design, and automatic interlocks.
| Hazard | Traditional Risk Level | New Equipment Risk Level | Mitigation |
|---|---|---|---|
| Alcohol vapor release | High | Low | Closed chamber and negative pressure |
| Liquid spill | Medium | Low | Sealed reservoir and level sensors |
| Fire or explosion | High | Low | Explosion-proof design and concentration monitoring |
| Operator exposure | High | Low | Automated cycle and filtered exhaust |
| Mold damage | Medium | Low | Spray instead of immersion |
| Waste overflow | Medium | Low | Weighing sensor and automatic stop |
The safety index can be expressed as a weighted sum:
$$S = \sum_{i=1}^{n} w_i r_i$$
where \(w_i\) is the weight of hazard \(i\), and \(r_i\) is its residual risk. In my assessment, the new equipment reduces \(S\) by more than 70% compared with the open immersion method. This is a direct benefit for any production environment that uses precision casting resin molds.
Solvent Lifecycle and Cost Considerations
The solvent lifecycle is managed through filtration, cycle counting, and concentration monitoring. The number of cleaning cycles per solvent charge can be estimated as:
$$N_c = \left\lfloor \frac{V_{solvent}}{V_{batch}} \right\rfloor$$
where \(V_{solvent}\) is the usable solvent volume and \(V_{batch}\) is the average solvent consumed per cleaning cycle. For a 30 L reservoir and an average consumption of 2–3 L per cycle, the equipment can perform at least 10 cycles before replacement or major filtration. This is a major improvement over the open tank, which may require frequent replacement because dissolved resin accumulates.
The total cost of cleaning can be modeled as:
$$C_{total} = C_{solvent} + C_{energy} + C_{labor} + C_{maintenance}$$
Although the initial equipment cost is higher than a simple open tank, the operating cost is lower because solvent consumption is reduced, labor time is shorter, and waste disposal volume is smaller. For precision casting production, the reduction in mold deformation and rework also contributes to cost savings.
| Cost Factor | Open Immersion | New Spray Equipment | Qualitative Effect |
|---|---|---|---|
| Solvent consumption | High | Low | Lower recurring cost |
| Labor time | High | Low | Higher productivity |
| Waste disposal | Large volume | Reduced volume | Lower environmental burden |
| Mold rework | Higher risk | Lower risk | Better precision casting yield |
| Energy use | Ventilation continuous | Adaptive ventilation | Lower average power |
Maintenance and Reliability
I designed the equipment for maintainability. The filter is accessible from the front panel, the solvent reservoir has a drain port, and the waste collector can be removed without tools. The HMI displays maintenance counters for the pump, fan, valves, and filter. A typical maintenance schedule is shown below.
| Maintenance Item | Interval | Check | Action |
|---|---|---|---|
| Alcohol filter | Every 20–30 cycles | Pressure drop | Replace or clean |
| Air filter | Every 50 cycles | Visual contamination | Replace |
| Nozzles | Every 100 cycles | Spray pattern | Clean or replace |
| Level sensors | Monthly | Signal continuity | Clean and calibrate |
| Weighing sensor | Monthly | Zero and span | Calibrate |
| Fan and encoder | Quarterly | Speed feedback | Inspect and lubricate |
| Explosion-proof enclosure | Quarterly | Seals and glands | Inspect and tighten |
Discussion of Precision Casting Quality
In precision casting, the resin mold is a sacrificial pattern. Its dimensions must be accurate, its surface must be smooth, and its burnout must leave minimal residue. Residual liquid on the surface can cause several defects. If uncured resin remains, it may cure unevenly during post-exposure, creating a skin that is thicker than intended. If alcohol remains in internal cavities, it can evaporate during burnout and create gas pockets. If the mold is soaked for too long, the walls can swell and distort. My equipment addresses these issues by using short, controlled spray cycles, directed air blow, and negative-pressure extraction.
The cleaning quality can be related to the final casting surface roughness. I define the surface quality index as:
$$S_q = \frac{1}{A} \int_A z^2 dA$$
where \(z\) is the local surface deviation. Lower \(S_q\) indicates a smoother surface. In my tests, spray-cleaned precision casting molds produced lower \(S_q\) values than immersion-cleaned molds because there was no redeposition of dissolved resin and no brush damage. The internal cavities were also cleaner and drier.
| Quality Indicator | Spray Cleaning | Immersion Cleaning | Precision Casting Impact |
|---|---|---|---|
| Surface residue | None visible | None visible | Both acceptable at macro scale |
| Internal cavity | Dry and clean | Liquid may remain | Spray reduces gas defects |
| Dimensional change | Very low | Higher | Spray preserves precision casting tolerance |
| Hardness change | Negligible | Softening observed | Spray protects thin walls |
| Repeatability | High | Low | Spray supports production consistency |
Comparison with Alternative Cleaning Approaches
I also considered other cleaning approaches, including ultrasonic cleaning, vacuum-assisted cleaning, and manual wiping. Ultrasonic cleaning can be effective, but it requires immersion in a tank, which reintroduces solvent volume and thin-wall softening risks. Vacuum-assisted cleaning can remove liquid from internal channels, but it does not by itself dissolve cured or semi-cured resin. Manual wiping is flexible but inconsistent and labor-intensive. The spray-and-blow method provides a balanced solution for precision casting resin molds because it combines dissolution, mechanical removal, and drying in a closed environment.
| Method | Solvent Volume | Thin-Wall Risk | Automation Potential | Suitability for Precision Casting |
|---|---|---|---|---|
| Open immersion | Very high | High | Low | Limited |
| Ultrasonic immersion | High | High | Medium | Limited |
| Manual wiping | Low | Medium | Low | Operator dependent |
| Vacuum assisted | Medium | Low | Medium | Partial |
| Closed spray and blow | Low | Low | High | Excellent |
Implementation Experience and Practical Observations
During implementation, I observed that nozzle placement is critical. If the spray is too concentrated, it can deform thin sections. If it is too diffuse, it leaves residual resin in corners. I therefore used a combination of fan nozzles and cone nozzles. Fan nozzles provide broad coverage, while cone nozzles provide localized impingement. The HMI recipe allows different zones to be activated in sequence, which prevents pooling and reduces solvent consumption.
I also observed that air blow is as important as solvent spray. After the final rinse, residual alcohol must be removed from internal cavities. If alcohol remains, it can soften the resin or create vapor during post-curing. The compressed-air blow step removes liquid from blind holes and channels. The negative-pressure ventilation then removes the vapor. This sequence is especially important for precision casting molds with complex internal geometry.
The weighing sensor on the waste collector proved to be a simple but effective safety feature. In open cleaning, waste liquid can overflow without warning. In the new equipment, the PLC continuously monitors the mass and stops the cycle before the collector reaches 90% capacity. This prevents spills and keeps the workspace clean.
Scalability and Future Development
The platform can be scaled in several ways. The chamber size can be increased for larger precision casting molds. The number of nozzle zones can be expanded for complex geometries. The solvent reservoir can be duplicated or connected to a central supply for higher production volumes. The PLC program can be extended with recipe management, data logging, and remote monitoring. Because the core architecture is modular, these changes do not require redesigning the entire system.
In the future, I plan to add machine vision for automatic residue detection. A camera could inspect the mold surface after cleaning and compare it with a reference image. If residue is detected, the PLC could run an additional spray cycle. I also plan to add solvent quality sensing based on refractive index or UV absorbance. This would allow more precise determination of when the alcohol must be replaced. Finally, I plan to integrate the cleaning equipment with the SLA printer and post-curing station so that precision casting resin molds move through a continuous digital workflow.
| Future Feature | Function | Expected Benefit |
|---|---|---|
| Machine vision inspection | Detect residual resin automatically | Adaptive cleaning for precision casting |
| Solvent quality sensor | Measure dissolved resin concentration | Optimize solvent replacement |
| Remote monitoring | Track cycles and alarms | Improve maintenance planning |
| Automated loading | Robot transfer of molds | Higher throughput for precision casting |
| Digital recipe library | Store parameters by mold type | Faster setup and better repeatability |
Conclusion
I have developed a high-performance cleaning equipment for removing residual liquid from precision casting resin molds produced by SLA additive manufacturing. The equipment replaces open immersion with a closed spray-and-blow process, uses quantitative solvent supply and recovery, and integrates PLC control with HMI visualization. The design meets the requirements of low surface residue, small dimensional deformation, high safety, and easy operation. In my tests, the new method reduced alcohol consumption by about 90%, shortened cleaning time by more than half, prevented wall softening, and maintained the dimensional accuracy required for precision casting.
The core contributions of this work are the integration of cleaning, ventilation, solvent recovery, and safety monitoring into one platform, and the demonstration that spray cleaning can outperform traditional soaking for thin-wall precision casting resin molds. The equipment is suitable for production environments where precision casting quality, operator safety, and solvent efficiency are critical. I believe this approach can be extended to other resin mold applications and can become a standard post-processing step for precision casting workflows based on SLA additive manufacturing.
In summary, the developed equipment provides a practical, scalable, and safe solution for residual liquid removal on precision casting resin molds. It addresses the main weaknesses of open immersion and manual cleaning, and it creates a repeatable process that supports high-quality precision casting production.
