Design Decisions That Quietly Double Your Casting Cost

There is a well-established principle in manufacturing cost engineering: 70–80% of a product's cost is determined by the time the design is frozen. For sand castings specifically, the proportion is even higher - because the geometry of the part directly determines the foundry's process cost, not just the material cost.

Procurement managers rarely have visibility into design decisions. But they absorb their consequences in every quote they receive. This article explains five design decisions that most significantly affect casting cost - not to turn procurement into engineers, but to give procurement the vocabulary to ask the right questions before tooling is committed.

Note: this article discusses cost implications only. CastCalc's design observations are advisory - any design changes must be validated by the responsible engineering team for functionality and safety.

1. Parting line placement

Every sand casting is produced in a two-part mould (cope and drag) that splits along a parting line. Where the designer places that parting line determines the complexity of the mould, whether cores are needed for certain features, and how much flash (excess material at the parting line) needs to be ground off.

A parting line placed at the maximum cross-section of the part - the natural split - allows the pattern to be extracted from both halves of the mould without undercuts. A parting line placed elsewhere often requires either cores (which add significant cost) or split patterns (which add tooling cost and increase cycle time).

The cost impact of a suboptimal parting line choice ranges from modest (if it just adds grinding time) to substantial (if it forces an additional core). On a medium-complexity hydraulic housing, a parting line revision that eliminates one core can reduce part cost by 8–15%.

2. Wall thickness uniformity

Uniform wall thickness is one of the most important principles of casting design - and one of the most frequently violated. When wall thickness varies significantly across a part (for example, a thin flange connected to a heavy boss), the different sections solidify at different rates. The heavier section solidifies last and can develop internal shrinkage porosity as it draws metal from the surrounding area.

To compensate, foundries add feeders (risers) - reservoirs of liquid metal positioned over the heavy sections to feed the shrinkage as it occurs. Feeders add metal consumption (which is then cut off and recycled, but the melt cost is real), add cycle time, and add finishing cost (the feeder root must be removed and ground).

A design that transitions abruptly from 8mm to 25mm wall thickness may require two or three feeders. A redesign that smooths the transition with a fillet or taper, or that relocates the heavy section, may eliminate feeders entirely. The yield improvement (less metal poured per finished kilogram) and the reduction in finishing cost can together save 10–20% on a heavily-fed part.

3. Internal cores and cavities

Any internal passage or cavity in a casting that cannot be formed by the mould halves alone requires a core - a pre-formed sand insert that defines the internal geometry and is destroyed during knockout after solidification.

Cores are expensive in proportion to their complexity. A simple cylindrical core for a straight bore is relatively cheap. A complex core for a curved internal passage with multiple entry points can cost more than the mould itself. Core cost includes: sand and binder, corebox tooling (often a separate investment), core machine cycle time, core assembly into the mould, and knockout and cleaning after casting.

Design decisions that affect core count and complexity include: whether internal passages are straight or curved (curved requires more complex - and expensive - coreboxes), whether multiple internal features can share a single core, and whether an internal passage is truly necessary in cast form or could be machined post-casting.

For a hydraulic manifold body with five internal passages, the difference between a design requiring three cores and one requiring five cores can represent 20–30% of total casting cost. This is rarely visible to procurement without a cost model that separates core cost from moulding cost.

4. Surface finish and tolerance specifications

As-cast surface finish on sand castings is approximately Ra 12–25 µm. Tighter surface finish requirements can be achieved by using finer sand, applying resin coatings to the mould, or specifying a different moulding process - all of which add cost.

More significantly, tight dimensional tolerances on as-cast features require more precise process control, slower cycle times to allow better control of solidification, and often higher scrap rates. Tolerances that are achievable in machining but specified as cast tolerances add substantial cost for small functional benefit.

The common mistake is applying general drawing tolerances (often ISO 2768-m or similar) to all features of a casting, including surfaces that will never be referenced or machined. Foundries quote against the stated tolerance, not against the functional requirement. A drawing that specifies ±0.3mm on a non-functional surface that could reasonably be ±1.5mm adds cost without adding value.

5. Material grade over-specification

Iron casting material grades span a wide range of mechanical properties and casting difficulty. Grey iron grades (GJL) are the most castable - they have good fluidity, low shrinkage tendency, and are forgiving of wall thickness variation. Ductile iron grades (GJS) require magnesium treatment of the melt, more careful process control, and are more sensitive to cooling rate variation.

The cost premium for ductile iron over grey iron on equivalent parts ranges from 15–30%, depending on grade, section size, and foundry. This premium is fully justified when the mechanical properties of ductile iron (tensile strength, elongation, impact resistance) are required. It is not justified when grey iron would meet the functional requirements.

Over-specification of material grade is common in two situations: when a design is based on a template that specified ductile iron for a previous application with higher loads, and when engineers specify ductile iron as a precaution against potential stress without a specific analysis confirming it is necessary. Both are worth revisiting with the engineering team - not to compromise function, but to ensure the specification matches the actual requirement.

Similarly, higher grades within the ductile iron family (GJS 600-3 versus GJS 400-15, for example) add cost - both through alloying and through the more controlled process required to achieve the specified properties. Specifying the minimum grade that meets the functional requirement is the correct engineering approach, and one that procurement can usefully prompt even without deep materials knowledge.

What procurement can do

The goal is not for procurement to redesign parts. It is to create a checkpoint - ideally at the point of design freeze, before tooling investment is made - where a casting cost expert reviews the design for obvious cost-drivers that can be addressed without compromising function.

The questions to ask the engineering team at this checkpoint are simple: Is the parting line placed at the natural split? Are wall thicknesses reasonably uniform? Can the number of cores be reduced? Are all tight tolerances functionally justified? Is the material grade the minimum required? Five questions, asked before tooling commitment, that can save 15–30% on casting cost for the life of the part.