The choice between a forged and a cast gear blank is decided long before the first tooth is cut, and it quietly sets the ceiling on strength, fatigue life, inspectability and unit cost for everything that follows. Two workpieces of the same drawing can behave like different components once they enter service, and the reason is almost always the way the blank was made. Forging deforms a solid billet plastically under hammer or press, crushing voids and aligning the grain along the part contour. Casting pours molten steel into a mould and lets it solidify in place, freezing whatever porosity, segregation and inclusion the solidification front leaves behind. Each route has a legitimate engineering niche, and the right answer depends on load severity, impact duty, part size, geometry complexity, batch volume and the inspection budget the customer is willing to pay for. This article walks through the metallurgical differences, the way non-destructive testing reads each route differently, two reference steel grades that sit on either side of the line, the cost trade-off, and the specification a buyer should put on the inquiry.
At the metallurgical level, the two routes start from the same idea but diverge almost immediately on density and grain structure. A forging begins as a continuously cast ingot or billet, which already contains centre-line porosity, segregation and non-metallic inclusions. The forging press or hammer then squeezes that billet in two or three directions, closing the internal voids, breaking up the cast dendritic structure and forcing the steel to flow around the die cavity. The result is a wrought microstructure with a fibre-like grain flow that follows the external contour of the part, so the flow lines run parallel to the tooth root stress path rather than being cut through by machining. Density is close to theoretical, typically above ninety-nine percent, and the directional toughness of a properly forged blank is noticeably better than that of a cast section of the same nominal chemistry. A casting, by contrast, solidifies from liquid. Shrinkage, gas pores, inclusions of slag or sand, and intergranular segregation are frozen into the section exactly where the last liquid metal turns solid. The grain is as-cast, coarse dendritic, and there is no deliberate fibre. Density is lower, and the mechanical properties in a heavy section are dragged down by whatever defects survived solidification, which is why the same chemical specification on a casting delivers lower guaranteed strength and fatigue limits than on a forging.
The difference in internal quality shows up immediately when the blank goes in for non-destructive testing, and the acceptance logic is not the same. Ultrasonic testing on a forging reads a clean, fine-grained matrix and looks for laminar flaws, internal cracks opened up during quenching, and forging folds that have been left untrimmed in the web or at a boss. Industrial practice grades forgings by the size and number of allowable reflectors in each depth zone, and the echo amplitude is compared against flat-bottom holes of known diameter. Because the matrix is fine, the background noise stays low and small reflectors are resolvable. On a casting, the same ultrasonic probe sees a coarse dendritic structure that scatters the sound already at the microstructure level, raising the noise floor and masking small indications. Castings are instead assessed for shrinkage cavities, piping, slag inclusions and hot tears, which tend to be larger in size but lower in number, and acceptance is usually based on radiographic or volumetric ultrasonic standards that permit larger local indications in less-stressed regions. In practice, this means a casting almost always carries a stricter visual, dimensional and NDT acceptance clause per kilogramme than a forging of the same size, because the supplier cannot assume the internal soundness the forging press has already delivered. Magnetic particle and penetrant testing on both routes then look for surface-breaking defects: laps and quench cracks on forgings, cold shuts and sand inclusions on castings.
Two grades illustrate the engineering split neatly. 42CrMo4, delivered as a quenched and tempered forging, is the workhorse of high-stress gear rims, pinions and shafts. After normalising, quenching in oil and tempering, it reaches a tensile strength in the nine hundred to one thousand one hundred megapascal range on suitably sized sections, with good notch toughness and a fatigue limit that supports case-hardened or induction-hardened tooth profiles. It is the default choice for gears that see shock loadings, reversing duty, heavy pitch-line pressures or thin-walled rims where a void would be catastrophic. On the casting side, the cast equivalent referenced under the DIN EN 10293 and ISO 3755 family of cast steel grades delivers a lower but still useful strength on heavier sections that would be uneconomic or impossible to forge. It is the standard answer for very large gears, ring gears with complex internal ribs, asymmetric housings, or blanks that carry a lot of integral geometry such as bolted flanges and lightening pockets. The design rule of thumb is simple: if the tooth root stress and the contact stress sit in the upper third of the material's allowable, or if the duty includes frequent shock, choose the forged route in 42CrMo4; if the gear is large, the geometry is near-net-shape, loads are low to medium and the cost of a forging die does not pay back, specify the cast equivalent and accept the lower guaranteed properties with tighter NDT.
Cost trade-offs follow from the two routes' fixed and variable components. A forging needs closed-die tooling, which is amortised over the batch, and it leaves generous machining allowances on the flange, the bore and the tooth faces because the as-forged surface is not accurate enough to run on. Material utilisation on a typical closed-die gear blank is around eighty percent, with the rest going to scale, flash and machined-off surplus. A casting needs a pattern and sometimes cores, which are cheaper than forging dies, and the geometry can be poured close to near-net shape, so the machined allowance is thinner and the chip volume smaller. On a large, complex blank that would otherwise require welding up ribs and flanges from plate, the casting often wins on weight removed and machine time saved. The crossover is empirical: for small, round, symmetric blanks up to roughly a few hundred kilogrammes, forgings from standard die sets are cheap and the tooling question disappears. Once the blank becomes large, asymmetric, ribbed or geometrically unique, the fixed cost of a closed forging die pushes the unit price up, and the casting route wins unless the duty cycle forces a forged microstructure. Batch size shifts the balance again: a one-off gear for a replacement winch often justifies a casting even when a forging would be stronger, because the tooling amortisation on a single piece is too small to matter in either direction.
When the route is chosen, the inquiry to the supplier should carry a short written specification so that what arrives matches what was intended. State the steel grade and the standard it is ordered to, for example 42CrMo4 to EN 10083 for a forging or the cast equivalent to EN 10293 for a casting. State the required heat treatment, delivery condition and hardness range on the tooth and on the hub, and whether tensile and impact tests shall be cut from an attached test coupon or from an extension on the blank itself. State the NDT acceptance: ultrasonic testing to a specified reflector class for the forging, radiographic or volumetric ultrasonic inspection to a specified grade for the casting, plus magnetic particle or penetrant inspection on all machined surfaces after heat treatment. State the drawing tolerances, the marking requirements, and whether material certificates to EN 10204 3.1 or equivalent are required. Finally, state the batch size and the required delivery schedule, because those two numbers decide whether the supplier reaches for a standard die, a new pattern, or a welded fabrication as a fallback. A blank specified on these six points is one the supplier can quote honestly, test against and deliver without reopening the metallurgy conversation after the order is placed.