Forged vs Machined from Billet: The Strength Argument Is Weak. The Arithmetic Is Not.

Forged vs Machined from Billet: The Strength Argument Is Weak. The Arithmetic Is Not. Ask ten suppliers why a part should be forged instead of machined from billet and nine will say the same thing: forging is stronger. It sounds right, it is repeated everywhere, and in the comparison buyers are actually making it is mostly false. A die forging of 7075-T6 and a piece of 7075-T6 plate land in the same tensile band. What forging changes is not the strength of the alloy. It is how much of the bar you had to buy, which way the grain runs, and whether the part through-hardens when you quench it. This guide separates the three claims, shows the arithmetic behind each, and gives you the point at which the die stops paying for itself. The decision matters because the two routes commit you to different things. Machining from billet commits you to chip volume and machine hours. Forging commits you to a die, a lead time, and a set of geometric constraints you will live with for the life of the part. Get the first one wrong and you overpay per piece, quietly, for years. Get the second one wrong and you write off tooling and start over. First, fix the comparison: it is forged vs wrought, not forged vs cast Most of the strength claims in circulation come from a comparison nobody is making. When a forging supplier publishes a table showing forgings beating the alternatives by 26 percent on tensile strength and 37 percent on fatigue, the baseline is almost always a casting. That comparison is real and it is not close. Cast aluminum contains gas porosity and shrinkage voids; forging closes them. But you are not choosing between a forging and a sand casting when your quote comes back. You are choosing between a forging and a part cut from wrought plate or bar. That distinction is the whole argument, so it is worth being precise about what “billet” means. Aluminum leaves the smelter as an ingot. It is then either poured into a mold, or hot-rolled and extruded into plate, bar and tube. That rolling and extrusion step is the quiet one that does most of the metallurgical work: it compresses the metal, refines the grain and closes the voids that casting leaves behind. By the time stock reaches a machine shop as “billet”, it is dense, uniform and predictable. It is already a wrought product. So a machined-from-billet part is not an inferior version of a forging. It is the same wrought material, cut to shape instead of pressed to shape. The correct question is not “which is stronger” but “what does pressing buy that cutting does not”. What the numbers actually show Here is the same alloy and temper, measured both ways. Die forging properties are longitudinal, tested to ASTM B557 for tension and ASTM E466 for fatigue at 107 cycles; wrought values are the standard published figures for plate and bar in the same tempers. Property6061-T6 wrought plate/bar6061-T6 die forging7075-T6 wrought plate/bar7075-T6 die forging Ultimate tensile strength310 MPa290–330 MPa572 MPa550–590 MPa Yield strength (0.2%)276 MPa240–290 MPa503 MPa480–520 MPa Fatigue strength (107 cycles)96 MPa95–115 MPa159 MPa150–180 MPa Elongation12–17%10–18%8–11%8–12% The tensile ranges overlap almost completely. Fatigue is the one row where forging shows a consistent, if modest, edge. Values are typical, not design allowables — use MMPDS or the applicable material specification for anything load-bearing. Read that table carefully, because it contradicts most supplier marketing. On tensile and yield, the two routes are indistinguishable within the spread of normal production. On fatigue, the forging band sits roughly 5 to 10 percent above the wrought figure at the midpoint. That is a real gain, and for a part that sees millions of load reversals it is worth having. It is not the 30 or 40 percent that circulates on supplier forums. The spread in published claims confirms this. Depending on which supplier datasheet you read, forging improves 6061-T6 tensile strength by 6 percent, or 7075-T6 by 11 percent, or “up to 26 percent” — and one widely circulated figure claims forged parts survive six times more stress cycles than machined ones. A 6 percent claim and a 500 percent claim cannot both describe the same physical effect. The conservative numbers are the ones that survive contact with the property table above. There is a second-order point hiding in that table, and it is more useful to a buyer than the forging question itself. Machined 7075-T6 billet, at 572 MPa, beats forged 6061 outright by a wide margin — and 6061 is the alloy most often proposed for forging because it flows well. If your part is strength-limited rather than fatigue-limited, changing the alloy and temper moves the dial far further than changing the process. Sort out the alloy first, then argue about the route. Our aluminum machining guide covers the grade-by-grade trade-offs, and the materials overview puts them next to the other families. What forging actually buys: three things, only one of which is about strength Strip out the marketing and the honest case for forging rests on three mechanisms. Only the first is mechanical, and it is directional rather than absolute. 1. Grain flow, which is about direction, not magnitude Wrought bar and plate have grain running in straight lines from the rolling or extrusion direction. Cut a hook, an arm or a crank from that stock and much of the geometry ends up loaded across the grain. Forging elongates the grains along the contour of the part instead, so the flow lines follow the load path and a crack trying to propagate has to cross them. The effect is real and it is anisotropic by nature. Research on hot closed-die forging of aluminum found that extruded forging stock has a flow stress roughly 20 percent higher in the extrusion direction than in the two perpendicular directions, which is why die filling and final grain orientation both depend on how the billet is presented to the die. The practical consequence for a buyer: grain flow is a property of the die design, not a free byproduct of the process. If your supplier has not shown you a flow-line sketch or a macro-etch of a sectioned first article, you are buying “forging” as a word rather than as an engineered grain structure. 2. Soundness, which matters most at the section size you cannot inspect Forging compresses internal voids closed. Casting leaves them. That is the source of the legitimate strength gap, and it is why forged parts are specified for landing gear, critical suspension members and pressure boundaries. It also has a finish-quality consequence that shows up on the shelf: forged aluminum has no gas porosity, so it anodizes far more uniformly than die-cast material. 3. Shape, which is where nearly all the money is This is the one that decides most quotes, and it is almost never the headline. A forging puts metal roughly where the finished part needs it. Machining from solid removes everything that is not the part. The difference between those two statements is the buy-to-fly ratio, and it is where the arithmetic lives. Buy-to-fly: the number that decides the quote Buy-to-fly is the weight of raw material you purchase divided by the weight of material that ends up in the finished part. A ratio of 5:1 means four fifths of what you paid for leaves the machine as chips. Published comparisons across aerospace manufacturing routes put the spread like this: RouteBuy-to-fly ratio Machining from forged block30:1 Machining from sections or plate12:1 Die forging (near-net shape)8:1 Form casting1.4:1 Pressure die casting1.2:1 Additive manufacturing1.2:1 Source: buy-to-fly comparison table published in Open Engineering (De Gruyter), “Implementation of high speed machining in thin-walled aircraft integral elements”, 2018. Figures are aerospace structural components — the spread is narrower for compact, chunky parts. The first two rows are worth dwelling on: they are the same process with different starting stock. Machining from a plain forged block is the worst route in the table at 30:1 — you are buying a rectangular solid and cutting most of it away. That is the configuration that gives machined-from-solid its reputation for waste, and it is the reason aerospace buyers push suppliers toward near-net preforms. A documented case makes the scale concrete. An aluminum main undercarriage fitting, machined from slab stock, started as 221 kg of material and finished at 12 kg — a ratio of 18.4:1. Re-engineered as a multi-stage closed-die forging, the same part came off a 50 kg billet at roughly 4:1. The team behind the work also noted a metallurgical benefit that the cost saving tends to obscure: machined-from-slab parts showed inferior properties precisely because they lacked continuous grain flow through the fitting’s load path. Translate that into a quote. Closed-die forging typically converts 70 to 85 percent of the input billet into parts, with the rest going to flash and trimming. Machining from solid on a structural aluminum part routinely removes 50 to 70 percent, and on a thin-walled aerospace-style component it removes far more. You pay for the removed material twice: once when you buy it, and again in machine hours, tool wear and chip handling while you remove it. Our breakdown of what drives machined part cost in China shows how dominant that second payment becomes. There is a floor on how much this matters. Chips are not landfill. Aluminum swarf is recycled at very high rates and carries real scrap value, so the material term is partly recovered. The machine-hour term is not. A near-net blank does not just save metal; it removes the roughing operations entirely, and roughing is where the hours go. That is the mechanism behind the widely cited estimate that near-net-shape starting stock cuts raw material cost by 15 to 35 percent on complex aerospace forgings compared with machining from solid — and it understates the saving, because it counts the metal and not the spindle time. The crossover: what the die has to earn back For a sourcing engineer. Everything above argues for forging. The die argues against it. Closed-die aluminum tooling runs from roughly $15,000 to $150,000 depending on size and complexity, with die life between 10,000 and 100,000 shots. That is a fixed cost you pay before the first good part, and it has to be recovered out of the per-piece saving. Published break-even figures vary enormously, and the variation is itself informative: 500 to 1,000 pieces — the point most closed-die specialists quote as the floor below which the die cost per part is too heavy. 1,000 to 50,000 units — the band where forging is described as the economic choice over machining from bar. 100 to 200 pieces — quoted by an Indian forger comparing closed die against open die plus machining, on a local cost base. Above roughly 1,000 parts per year — the threshold one aluminum forging guide gives for beating machining from solid bar. Do not pick a number from that list. The crossover moves with four variables, and you can estimate it yourself in about ten minutes: Die cost, quoted for your geometry, not a rule of thumb. Material saved — the weight difference between the finished part and the stock each route starts from, times your alloy price. Machine hours saved — roughing passes eliminated, times your shop’s hourly rate. Program volume — total parts over the life of the design, not one order. Then divide the die cost by the sum of items two and three. If that gives you fewer parts than you will actually build, forging pays. If it gives you more, it does not, and no amount of fatigue performance changes the answer for a part that is not fatigue-critical. One trap worth naming: the per-piece saving shrinks as your part gets chunkier. Buy-to-fly favors forging most when the finished part occupies a small fraction of its bounding box — arms, levers, yokes, brackets with thin webs between heavy bosses. If your part is essentially a rectangular block with holes in it, there is no buy-to-fly argument, because a saw-cut blank already costs almost nothing to prepare. For that geometry, die casting is the process that changes the economics, not forging. Four constraints a forging imposes on your design, and what each one costs in CNC For a manufacturing engineer. Choosing forging is not a free substitution. It hands the CNC programmer a different set of problems, and most of them are locked in before the die is cut. ConstraintTypical valueWhat it means downstream Draft angle5–7° external, 7–10° internal, up to 15° in deep pocketsNo truly perpendicular walls on as-forged surfaces. Any face that must be square to a datum has to be machined. Machining allowance1–3 mm per surfaceForging gets you close, never there. Budget a finish pass on every functional face. As-forged tolerance±0.3–1.0 mm hot; ±0.1 mm cold on simple shapesPractical rule: if the gap between as-formed and required tolerance exceeds 0.2 mm, you are paying for secondary machining. Anything locating tighter than about ±0.15 mm gets cut. Fillet radiiGenerous — sharp internal corners kill diesDesign out sharp internal corners. This also helps you machine them, since a sharp corner needs a small cutter and a slow feed. The parting line deserves a line of its own. It is where the die halves meet, and it does three things at once: it puts flash there, it determines where you will need a trimming operation, and it interrupts grain flow exactly where the metal changed direction last. Put it at the largest cross-section and expect to machine across it. A supplier who cannot tell you where their parting line sits, and why, has not finished designing the die. There is also a scheduling consequence. Tooling lead time for closed-die forging runs 6 to 12 weeks, against days for a part that goes straight onto a mill. One shop’s account of a 50-piece fuel rail order puts the practical gap at ten days versus six weeks. If your design is still moving, that lead time is not just a delay — it is a commitment to a frozen geometry, and every subsequent change either cuts new tooling or gets absorbed by the machining operation. Thick sections: the one place the argument genuinely reverses So far the metallurgical case for forging has been modest. Section thickness is where it stops being modest, and it is the least discussed of all the mechanisms. Heat-treatable aluminum alloys gain their strength from solution treatment and a fast quench. The quench is the problem. Metal can only be cooled as fast as heat can travel from the core of the section to the surface, so a thick section through-hardens poorly no matter what route you used to make it. Measured cooling rates on 7000-series sections make the penalty explicit: Section thicknessSurface cooling rateCore cooling rateExpected core hardness drop 0.5 in300+ °F/s250 °F/sMinimal 2.0 in250 °F/s100 °F/sModerate, 5–10% 4.0 in150 °F/s40 °F/sSevere, 15–25% 6.0 in and above100 °F/s