Somewhere between the design intent and the shop floor, a lot of casting projects go sideways on one small block of text: the heat treat callout. Engineers who agonize over wall thicknesses and fillet radii will sometimes write “T6 per spec” on a drawing and consider the metallurgy handled. Other engineers go the opposite direction and specify so much testing and documentation that the part costs more to certify than to cast.
Both mistakes are expensive, and both are avoidable. Getting the callout right matters because aluminum heat treating is where a casting earns its mechanical properties. The as-cast part is only the raw material for what it becomes after solution treatment, quenching, and aging. The drawing tells the foundry what the part needs to become, and a vague or overloaded callout leaves the foundry guessing or charging for requirements the part will never use.
This article covers what belongs in a heat treat specification, what does not, and the gray areas where the right answer depends on the application.
Start With the Alloy, Because It Determines Everything
Before writing a single requirement, confirm the alloy can actually be heat treated, because not all of them can, and not all of them need to be.
The heat treatable casting alloys, 356, A356, 319, C355, and 206 families among them, respond to solution treatment and aging because their alloying elements dissolve at elevated temperature and reprecipitate in a controlled way during aging. That precipitation is what builds strength and hardness beyond the as-cast condition.
Then there are the self-aging alloys. SR319 and 713 develop their properties at room temperature over time, no furnace required. Specifying a full solution treatment and artificial aging cycle on one of these alloys is a red flag on a drawing, and a good foundry will call it out. If you have specified a self-aging alloy, the heat treat callout may be limited to stress relief or nothing at all, and that is correct, not an omission.
The practical takeaway: the alloy selection and the temper selection are one decision, not two. Choosing an alloy because it pours well and then demanding the mechanical properties of a different alloy family is a specification that cannot be satisfied.
What Belongs on the Drawing
The temper designation. This is the core of the callout, and it should be specific. T4 means solution treated and naturally aged, which buys formability and moderate strength. T6 means solution treated and artificially aged, which maximizes strength and hardness. T5 and T51 skip the solution treatment and apply artificial aging directly from the casting or fabrication temperature, adding strength economically when full solution treatment is not needed. T7, T71, and T77 add stabilization, trading a little hardness for dimensional stability and improved corrosion resistance.
The mistake to avoid is treating these as interchangeable. A housing that must hold dimensions over years of thermal cycling may genuinely be better served by a stabilized T7 temper than by a maximum-hardness T6, even though T6 looks stronger on a datasheet. If you know the part’s failure mode is dimensional drift, specify for stability. If it is static load, specify for strength. If it is shock, pay attention to what you are giving up in ductility with each hardness step.
The governing specification. Name it. Commercial castings are commonly specified per ASTM B26 or Aluminum Association standards, and defense work often invokes MIL-A-21180, with class and grade called out. The specification determines what the heat treat cycle must achieve, how properties are verified, and what documentation accompanies the parts. “Heat treat per industry standard” is not a specification, and it will generate a clarifying phone call at best.
Mechanical property requirements, where they matter. If the design depends on minimum tensile strength, yield strength, or elongation, say so with numbers. This is what drives test bar requirements: tensile test bars poured with the production lot and processed through the same heat treat cycle are how the properties get verified. If your part carries real structural load, the test bars are not optional overhead. They are the proof.
Hardness requirements, carefully. Brinell hardness is a quick, cheap verification that a heat treat lot landed in the right neighborhood. It works well as a lot acceptance check. It does not work as a full substitute for tensile verification on critical parts, because hardness correlates imperfectly with strength and says nothing about ductility. Specify hardness as a screen, tensile as proof, and you get a sensible inspection regime at a sensible cost.
Special requirements that affect processing. If the part has tight dimensional tolerances on machined surfaces, say so, because quench distortion is real and the foundry can respond with fixturing, quench technique adjustments, or straightening operations. If the part must hold dimensions in service, that is the case for a stabilized temper. If there are areas where distortion is unacceptable, flag them. The heat treat department cannot accommodate a constraint it does not know about.
What to Leave Out
Process instructions. Solution temperature, soak time, quench medium, and aging schedule are the foundry’s domain, governed by the specification you named. Writing your own cycle on the drawing ties the supplier’s hands, and unless you have metallurgical staff who want to own the outcome, it adds liability without adding performance. Specify the outcome. Let the process engineers own the path.
Testing that duplicates itself. A common pattern is a drawing that calls for tensile bars, hardness tests, radiographic inspection, and penetrant inspection on every part, when the specification and the application justify a fraction of it. Each requirement is individually reasonable, and together they can double the part cost. Radiographic inspection belongs on parts where internal porosity creates a functional risk, like pressure containment. It is wasted money on a bracket whose failure would be visible and non-critical. Match the inspection regime to the consequence of failure, not to a template.
Properties the part does not need. Maximum hardness on a part whose load is static, maximum strength on a part whose real enemy is corrosion, and full solution treatment on an alloy that gets nothing from it are all examples of specification creep. Every property you demand is a cost, a lead time, and occasionally a trade you did not intend, because pushing hardness up pushes ductility down.
Ambiguity, in either direction. The two failure modes of a heat treat callout are underspecifying and overspecifying, and they fail differently. Underspecified parts arrive with properties nobody can verify. Overspecified parts arrive late and expensive, with a certification package nobody reads. Both trace back to the same root cause: nobody decided what the part actually needs to survive.
The Questions Worth Asking Before You Finalize the Callout
Three questions resolve most specification arguments before they start.
What is the failure mode? Fatigue, overload, corrosion, wear, and dimensional drift each point to different temper and property choices, and none of them is answered by defaulting to T6.
What happens if this part fails? The answer sets the inspection regime. A ground vehicle structural component and a commercial equipment cover should not carry the same certification burden.
How will the properties be verified? If the answer is “the foundry’s word,” either the part is non-critical or the drawing is missing a test bar requirement. If the answer is a stack of certificates nobody will read, the pendulum has swung too far the other way.
The Bottom Line
A heat treat specification is a conversation between the designer’s intent and the foundry’s process, conducted in advance, on paper. The good ones name the alloy, the temper, and the governing specification, demand verification proportional to the consequence of failure, and stop there. The bad ones either assume the foundry will figure it out or bury the part in requirements that exist to make the drawing feel rigorous. Spend the extra hour deciding what the part genuinely needs to survive, and the callout will mostly write itself. The foundries that see clean, deliberate specifications recognize them immediately, because they are rare, and they tend to treat the drawings that contain them with the same care.