Geometry Is the Real Decision Filter
The cleanest way to separate aluminum extrusion from casting is to stop treating them like interchangeable processes. They are not. Extrusion is a profile process. Casting is a volume process.
That distinction sounds simple, but it explains most of the real-world outcomes engineers care about: tooling cost, lead time, material waste, tolerance stack-up, finish quality, and even the number of secondary operations needed before a part can ship. The tooling delta is not the reason geometry matters; it is the evidence.
If the part is one cross-section pulled through length, extrusion is the default. If it is a volume with features in three axes, casting is the more honest fit.
Engineers who use a decision matrix usually start with one question: can the part be described as one cross-section repeated along a length? If the answer is yes, extrusion is already the natural candidate. If the answer is no, casting starts to look less like a backup plan and more like the process that fits the shape.
A Profile Wants Extrusion
A part that keeps the same section from end to end is telling you how it wants to be made. Window frames, heat sinks, structural rails, conveyor guides, LED housings, and battery tray members all share that same design logic: the geometry is long, consistent, and predictable.
That is why extrusion tooling is relatively inexpensive. A die only has to define one cross-section. Once the aluminum is pushed through it, the process can keep making that same shape over and over at different lengths. In practical sourcing terms, that means:
- lower tooling investment, often under $1,000 for simpler dies and far below casting mold costs
- faster die lead times, commonly a few weeks rather than months
- high material utilization, often around 85-92%
- linear tolerances around ±0.1 to ±0.3 mm for many profile dimensions
- good as-extruded surface quality for anodizing or powder coating
The part itself benefits too. A profile with a constant cross-section is not asking the process to improvise. Tolerances are easier to control, scrap stays lower, and secondary machining can often be limited to cut length, holes, or end features. A 600 mm LED heat sink is a profile problem. So is a curtain-wall mullion or a machine guard rail.
A Volume Wants Casting
Casting solves a different problem. It exists for parts that are not honest profiles: housings, bodies, brackets with bosses in multiple directions, parts with variable wall thickness, and components with internal cavities or enclosed flow paths.
Those shapes are difficult for extrusion because a die can only create the same section continuously. Once the design starts changing in three dimensions, the process has to rely on machining, welding, or assembly to recover the missing geometry. At that point, the part is no longer a clean extrusion candidate.
Casting accepts that complexity directly. The mold defines the full three-dimensional form, so ribs, pockets, mounting pads, and passageways can be integrated into a single piece. The price of that freedom is predictable:
- higher tooling cost, often well above $30,000 and sometimes much more
- longer lead times, often 8-16 weeks for production molds
- more material loss from runners, gates, and flash
- more variation from section thickness and shrinkage behavior
- material utilization that often lands closer to 60-80%
- more frequent post-cast machining on critical surfaces
A pump housing with a volute, threaded bosses, and a sealed cavity is a casting problem. So is a complex motor cover or a bracket that needs material to branch in several directions without assembly.
Why Cost Follows Geometry
The most useful way to think about cost is not as a separate criterion but as a consequence of shape.
If a part is a clean profile, extrusion wins on cost because the geometry is simple enough to support a simple tool. If a part is a 3D body, casting wins on feasibility because the geometry demands a full cavity. Every other variable sits downstream from that first decision.
A die only has to define one cross-section. A mold has to define a cavity, gates, vents, ejection, cooling, and shrink compensation. That is why the tooling gap can run from under $1,000 for simple extrusion dies to well above $30,000 for casting tools, with production molds sometimes far higher.
That is also why the wrong process becomes expensive in ways that are easy to miss on a quote sheet. A low per-piece price can hide a geometry mismatch that creates scrap, secondary machining, or assembly labor later. A higher tooling quote can still be the cheaper option if it buys the exact shape the product actually needs.
The right question is not which process is cheaper. It is which process matches the geometry with the fewest compensations.
When the Part Should Be Redesigned Instead of Reprocessed
Some projects never should have been framed as extrusion versus casting in the first place. They should have been framed as design-for-manufacture problems.
If a cast part could be split into an extruded profile plus machined ends, the redesign often saves more than negotiating raw material price or chasing a lower quote. If an extrusion candidate needs one local boss or one closed cavity, a small cast insert or a welded end feature may solve the issue without forcing the whole part into a mold.
That mindset matters because the CAD model can be deceptive. A beautiful 3D shape does not always mean the function needs a 3D manufacturing process. Sometimes the shape only looks integrated because no one has decomposed it into its functional geometry.
The Fast Test Engineers Use
A quick geometry-first review usually exposes the answer before any quote arrives.
- If the part can be drawn as one constant cross-section, extrusion is in play.
- If the part needs bosses, pockets, and walls that change direction, casting is in play.
- If the only variable is length, extrusion usually fits.
- If the part must exist as a three-dimensional body, casting usually fits.
- If the design can be split into a profile and a small machined feature, extrusion deserves a hard look before anything else.
That logic explains why the best selection framework always puts shape ahead of price. The geometry determines the tooling, the tooling determines the lead time, the lead time determines the launch plan, and the shape also drives how much machining, finishing, and scrap the project will absorb.
The Habit That Prevents Expensive Mistakes
The highest-leverage move is not finding a cheaper casting quote or a faster extrusion press. It is making the geometry speak first.
When the part is fundamentally a profile, extrusion usually delivers the cleanest path: lower tooling, better utilization, tighter control, and fewer surprises. When the part is fundamentally a body, casting earns its keep by making complex volume possible in one piece.
That single habit prevents the most expensive mistake in aluminum sourcing: paying for process capability you do not need, or denying it when the shape actually demands it.