METHOD: THE MATHEMATICAL FRAMEWORK

(From Theory to Industrial Reality: Algorithmic Decision Support Protocols)

JoseCast rejects the heuristic "trial-and-error" approach of classical foundry practices. The program is an advanced Mathematical Casting Framework that unifies thermodynamics, fluid mechanics, and linear algebra principles into a deterministic model.

Rather than statically using pure physics formulas designed for laboratory conditions, the computational engine at the core of the system calibrates them using "Dynamic Correction Functions" sensitive to mass, geometry, and chemical composition. Acting as the "First Line of Defense" prior to heavy Finite Element Analysis (FEA/Simulation), JoseCast breaks the Garbage-in, Garbage-out cycle.

The algorithmic depth of the modules forming the backbone of JoseCast is detailed below:

🔬 1. Thermodynamic Adaptation and the Enhanced Chvorinov Model

Solidification time in casting is not a simple geometric ratio. JoseCast transforms classical Modulus Calculation into a multi-dimensional thermal analysis.

  • Dynamic Chvorinov Exponent: Standard solidification theory utilizes the formula t = B(V/A)2. However, JoseCast does not leave the square coefficient (n=2) static; it adapts based on section thicknesses and thermal mass distribution. To compensate for corner effects, radius constrictions, and sand heat saturation, it dynamically shifts the n exponent between 1.5 and 2.2.
  • Niyama Criterion (Ny) Integration: To calculate the risk of microporosity (shrinkage) without waiting for full simulation, it runs a preliminary Ny = G / √Ṫ analysis in the background using the local temperature gradient (G) and cooling rate (Ṫ). Riser (feeder) modules are sized according to this preliminary criterion.
  • Ductile Iron Superheat Compensation: The endothermic nature of Magnesium modification and Silicon inoculation is integrated into the system. Depending on the variance in the Si ratio between 1.5% and 3.0%, an algorithmic additional Superheat value (ΔT) is assigned, eliminating pre-casting cooling risks.

⚙️ 2. Kinematic Flow Control: Campbell's Rules and Dampened Bernoulli

In-mold metal flow is not a standard hydraulic piping system. JoseCast does not directly use the pure Bernoulli equation (v = √(2gH)) for gating designs; Campbell’s fluidity rules and empirical dampening algorithms come into play.

  • Froude Criterion and Oxide Bifilm Control: The system limits the entry velocity of the liquid metal into the mold via the Froude number (Fr = v / √(g · h) < 0.5). This critical threshold theoretically prevents surface turbulence (vortexing) during pouring and, consequently, oxide bifilm (dross) inclusions that destroy the mechanical life of the part.
  • Reynolds and Head Reduction: Metallostatic pressure is dampened by in-mold friction and directional changes. JoseCast increases the pressure dampening coefficient, which is around 40% for a 100 kg part, to 75% for massive 3000 kg parts. This prevents sand erosion (sand wash) and guarantees a laminar flow regime.
  • Logarithmic Filling Anchors and Density Multiplier (fρ): To equalize the in-mold kinematic behaviors of different alloys, a dynamic kinematic viscosity correction is applied based on the reference density using the formula fρ = (ρref / ρ)0.35.

🧮 3. SLSQP Matrix Optimization: "Least-Cost" Charge Design

The Melt Charge module is the linear algebra hub where the system simultaneously manages industrial profitability and quality. Beyond basic Excel "Goal Seek" functions, it operates on the SciPy library-based SLSQP (Sequential Least SQuares Programming) algorithm.

  • Thermodynamic Mass Balance: Scrap, pig iron, and ferroalloy inputs are transformed into an n-dimensional matrix array. The algorithm optimizes not only the elements (C, Si, Mn, etc.) but also the targeted Carbon Equivalent (CE = %C + (%Si + %P) / 3) and the Degree of Eutectic Saturation (Sc) as distinct "Boundary Conditions."
  • Non-Linear Penalty Functions: By applying a mathematical "penalty" to any combination tending to drift outside the targeted analysis, the algorithm iteratively pulls the matrix's trajectory into tight tolerance bands.
  • Minimum Cost Optimization: Once all metallurgical conditions are satisfied, the system scans the available raw material inventory and calculates the optimum recipe that minimizes the furnace cost (Least-Cost). This radically reduces alloy waste in foundries and "correction" times at the spectrometer.

🛡️ 4. System Security and Isolation Architecture

JoseCast is built with a dual-layer architecture to ensure stability. While the "Outer Armor" (Launcher/Radar) system monitors processes within the operating system; the "Quarantine Zone" where calculations occur is completely isolated from external factors, providing 100% mathematical consistency and operational data security.

⚖️ Engineering Responsibility Framework (Disclaimer)

JoseCast is an analytical terminal calibrated to reduce theoretical design errors to zero, taking on the heavy matrix calculation load in design offices. However, the science of industrial casting depends on hundreds of chaotic variables—from molding sand parameters to furnace atmosphere, from refractory reactions to the kinematics of the pourer's ladle tilt. In this context, the data generated by the software is not an absolute "conclusion" but a reliable, thermodynamically perfected "Starting Point." The final production decision and manufacturing approval are always subject to the field experience, initiative, and professional judgment of the casting engineer using the software.

Algorithms solve the math, the engineer makes the cast.