Forged Vs Cast Valve Body: Size, Class And Cost Boundaries

"Is forged better than cast?" is the question that arrives with almost every small-bore enquiry, and it is the wrong question. Both routes produce bodies that meet the same pressure-temperature rating in ASME B16.34. Neither is stronger in the abstract. What actually differs is the kind of uncertainty you inherit, the size and class range in which each route is commercially available, and the quantity at which the tooling economics flip. A DN25 Class 800 gate valve body is forged because casting it would be absurd, not because forging is metallurgically superior. A DN400 Class 150 gate valve body is cast because a forging of that mass does not exist at a price anyone will pay.

This article draws the boundary properly. It separates the metallurgy from the sales language, then puts numbers on the overlap zone where the decision is genuinely open — roughly DN50 to DN150 — and gives the specification clauses that decide the outcome when price alone does not.

1. Two Routes, Two Different Kinds of Uncertainty

Casting turns liquid metal into a shape. The body is poured into a mould, solidifies, and is then machined on the surfaces that matter. The uncertainty this creates is internal and volumetric. Where a heavy flange hub meets a thin barrel, the thin section freezes first and the hub keeps drawing liquid; if the feeder cannot keep up, a shrinkage cavity opens in the last region to solidify. In austenitic stainless steels a second mechanism runs in parallel: hydrogen solubility drops sharply as the alloy freezes, gas is rejected at the advancing solid front, and where the shell closes faster than the gas can escape, gas porosity remains. Add sand or slag inclusion, hot tearing where a rigid mould refuses to yield to contraction, and cold shuts where two flow fronts meet without fusing, and you have the standard casting defect list.

Forging deforms solid metal. A billet is heated and worked between dies or platens, and the shape is produced by displacement rather than solidification. There is no shrinkage cavity to form in the part and no gas to reject, and the grain structure is refined and redirected to follow the contour, which is worth something on a body that carries a bending load between the pipe flanges and the bonnet. What forging does not do is purify the steel. A billet with a segregated, porous centreline stays segregated and porous in the middle if the reduction ratio is too low to close it, and a poorly controlled pass can fold the surface into a lap that behaves like a crack.

The practical consequence is that the two routes fail differently, and neither route is exempt from inspection. A casting fails by weeping through interconnected porosity that was always there. A forging fails at a lap, a seam or a billet defect, or at a weld if one was made. Anyone who tells you a forged body needs no volumetric examination is selling, not specifying.

2. The Code Boundary Is Really a Size Boundary

The standards have already made most of the decision for you. API 602 governs forged and bar-stock gate, globe and check valves up to NPS 4 (DN 100), and API 600 and API 623 govern cast steel gate and globe valves from NPS 2 upward. That overlap band of NPS 2 to NPS 4 is the only place where the choice is open by default; everywhere else, one route is the standard answer and the other is a special order.

Two details in that table trap buyers. First, Class 800 is not a B16.34 class at all; it is an API 602 rating that sits between Class 600 and Class 900 and exists only for small-bore valves. A DN25 Class 800 socket-weld gate valve and a DN25 Class 900 flanged gate valve are therefore not interchangeable even though the numbers look adjacent. Second, API 602 permits both forged and rolled bar stock bodies. Those are not the same product, and the difference is covered later in this article. For the product families on the forged side, the forged steel valve range shows the size and class envelope in practice; the cast side is covered by the cast steel gate valve range.

3. Why Cast Bodies Are Heavier: The Casting Quality Factor

ASME B16.34 does not treat a casting and a forging as metallurgically equivalent when it calculates minimum wall thickness. The wall equation carries a casting quality factor, and the values are blunt: 0.80 for a standard casting, 1.00 for a forging, and 1.00 for a casting that has been upgraded to premium quality by full volumetric examination. Because the factor sits in the denominator, an 0.80 factor demands roughly 20 to 25 percent more wall than a 1.00 factor for the same pressure, diameter and allowable stress.

That single coefficient explains a lot of what buyers notice on the loading dock. A standard-class cast body is thicker-walled than a forging of the same rating before any tolerance is added, and the tolerance is not trivial: a minimum wall is a minimum, so a foundry must pour nominal wall plus casting tolerance plus machining allowance to be sure of complying after clean-up. The forging starts from a solid billet with no draft angle and no as-cast skin, so its allowance is machining only.

Note the reversal in the fourth row, because it is the part that gets lost. In small bore the forging wins the weight argument comfortably: a 1 inch Class 800 forged gate valve is roughly one-third the length and a fraction of the mass of a cast Class 150 valve of the same nominal size. Above DN150 the logic inverts. A casting puts metal only where the pressure boundary and the flow path need it; a forging of the same size starts as a billet large enough to contain the whole body and machines most of that billet into chips.

4. What Each Route Must Prove, and What a Hydro Test Does Not Prove

This is where the specification is usually weakest. ASME B16.34 distinguishes a standard class from a special class, and the special class carries additional nondestructive examination. Buy to standard class and no radiography happens unless you write it into the purchase order, with a reference standard, a severity level and a coverage rate. "No porosity" is not a purchasable requirement; "ASTM E446, category gas porosity, severity 2, flange hub and seat pocket sections, 100 percent" is.

The last row deserves its own paragraph, because more procurement arguments start here than anywhere else. A hydrostatic shell test proves that the pressure boundary holds at test pressure, at ambient temperature, for the duration of the test. It does not prove the body is sound. A subsurface shrinkage cavity surrounded by sound metal passes that test without difficulty, then intersects the wetted surface once corrosion has removed a fraction of a millimetre, and weeps. In chloride service the void is worse than a void: it is a stagnant geometry that stabilises a crevice, so pitting starts there and grows inward. This is why a valve can pass API 598 in the shop and wet its flange hub in year two. The test procedure itself is covered in our API 598 pressure testing guide; the point here is that the shell test and the radiograph answer two different questions, and only one of them is about soundness.

5. The Weld Repair Clause Decides More Projects Than Price Does

Foundries repair castings. The question is never whether it happens; it is whether the repair is excavated to sound metal, welded to a qualified procedure under ASME Section IX, stress-relieved, re-examined by the same method that found the defect, and disclosed on a repair map tied to the heat number. Undisclosed repair in a pressure boundary is a different product from the one that was ordered, and it is invisible on a finished, painted, passivated body.

For most utility service a disclosed, qualified repair is acceptable engineering. For a growing list of duties it is not, and the list is where the forged route earns its premium:

• Sour service to NACE MR0175 / ISO 15156, where hardness control and heat treatment of a repair zone are difficult to guarantee and many operators prohibit pressure-boundary repairs outright.

• Hydrogen service, where the repair zone introduces a hardness and residual-stress discontinuity in exactly the region the material selection was trying to protect. Our hydrogen service valve specification guide treats repair policy as a material requirement, not a quality clause.

• Oxygen and high-purity service, where any cavity or repair is a contamination trap and a ignition site.

• Wet seawater and hypochlorite, where surface discontinuity is a crevice that the alloy's corrosion allowance was never meant to absorb. The same logic drives alloy choice in our duplex and super duplex selection guide for seawater.

Once a specification says "no pressure-boundary weld repairs", a standard casting cannot be certified as compliant without full radiography and a repair map, and even then you are relying on the foundry's disclosure. A forging has nothing to repair in the first place, apart from the weld ends if the body is a welded fabrication. That, more than any strength claim, is why DN150 Class 900 gate valves for sour service are routinely quoted as forged bodies at two to three times the cast price.

6. The Cost Model: Two Tooling Curves That Cross

Direct price comparison between a cast and a forged valve is meaningless unless the tooling is carried in the same arithmetic. The two routes have opposite cost structures. Casting carries modest tooling and a piece price that falls only slowly with volume. Closed-die forging carries heavier tooling and a lower piece price once the die exists. Open-die forging or machining from a forged block carries no tooling at all and the highest piece price.

Total cost for a quantity n is simply tooling plus n times piece price, so the crossover quantity is the tooling difference divided by the piece-price difference. Call the low-tooling route A and the high-volume route B:

n* = (T_B - T_A) / (P_A - P_B)

Below n*, route A wins on total cost. Above it, route B wins. The numbers below are indicative 2026 levels for carbon steel bodies, single-cavity tooling, FOB China, normalised to the same trim and the same test scope. They are close enough to reason with and must be re-quoted for any real project.

Read the table as a procurement tool rather than a price list. The DN100 case is the one that surprises people: for a spares order of 40 units, machining forged bodies from block costs less in total than paying for a pattern set, and it arrives in three to four weeks instead of the twelve to sixteen a new pattern and first article usually take. That is why a shutdown spare is so often a forged body even when the original installation was cast. At 300 units the same comparison reverses and the casting wins by a wide margin.

Pressure class moves the crossover hard, in a direction most buyers do not expect. As class rises, the cast piece price climbs faster than the forged one, because wall thickness, section size, radiography coverage and heat treatment all scale with class, while the cast tooling cost also rises. The result is that the quantity needed to justify a casting falls as class rises, and above about Class 900 at DN100 the forging is cheaper per piece before tooling is even considered.

The DN100 Class 1500 row is the honest summary of the whole argument. At that size and class the casting has become a thin-walled, high-integrity component that must be radiographed anyway, so its cost advantage has evaporated and its quality uncertainty has not. Below DN100, the same effect arrives earlier in class. Above DN150, the forging's piece cost has grown so far beyond the casting's that no realistic order quantity recovers it.

7. Where the Forged Route Wins Regardless of the Cost Sheet

Six situations override the arithmetic, and a specification that ignores them will produce a valve that is cheap and wrong.

• Small bore at high class. DN50 and below at Class 1500 and 2500 is forged territory. Casting a thin, complex, high-pressure small body reliably is a different manufacturing problem from casting a DN300 body, and the reject rate shows it.

• A no-repair policy. Sour, hydrogen, oxygen and high-purity service, as covered above. If the clause is in the spec, the route is decided.

• Weight and space on an offshore module. A compact forged body with socket-weld or extended butt-weld ends removes weight from the structure and length from the pipe run, and on a module those are the two currencies that matter.

• Thermal cycling and low temperature. Forged ASTM A350 LF2 bodies are the standard answer for small-bore cryogenic and low-temperature service, where a casting's section changes are stress raisers under repeated cooldown.

• Extended bodies, double block and bleed, and instrument manifolds. These are bars with cross-drillings and multiple seat pockets. Nobody casts them, because the geometry is a machining problem, not a moulding problem.

• Vibration and fatigue. Grain flow running continuously around the bore and into the bonnet thread region resists fatigue crack initiation in a way that an as-cast surface cannot be relied on to do.

8. Where the Cast Route Wins Regardless of the Cost Sheet

The reverse list is shorter but harder to argue with.

• DN150 and above. Forging press capacity and billet availability set a hard economic ceiling, and for most standard gate, globe and check bodies that ceiling is around DN150 at low class and DN80 to DN100 at Class 900 and above.

• Complex internal geometry. A full-port ball valve with trunnion pockets, a double block and bleed body with two seat bores and a cavity relief port, or a cage-guided globe body cannot be forged as a cavity. Machining those from a forged block would turn most of the billet into chips.

• Large flow capacity. Once the required Cv runs into five figures, the bore size itself decides the route. The relationship between Cv and required bore is covered in our Cv and Kv sizing calculation guide.

• Jacketed and special-alloy bodies. Steam-jacketed bodies and grades that are only produced as castings — ASTM A995 cast duplex and super duplex grades, for example — have no forged equivalent in the required size. Forged A182 F51 and F53 bar is available, but not in the mass a DN400 body needs.

• Cost per kilogram. Cast steel lands in the region of two dollars per kilogram of finished body at moderate size; forging conversion costs several times that, and the gap widens with mass.

9. What to Write in the Purchase Specification

Most of the failures attributed to "bad material" are failures of the purchase order. These ten clauses close the gap, and each one is short enough to paste into a datasheet.

10. How Each Route Fails: Read the Leak

When a body leaks, the location and the shape of the leak usually identify the route and the cause, and that determines whether you have a warranty claim or a specification gap.

The last row matters because route arguments are frequently used to explain a seat failure. A body that does not leak and a seat that does are two separate findings, and the second one is a lapping, alignment or debris question rather than a metallurgy question.

11. The Working Rules

1. Below DN50, specify forged or bar stock to API 602 and stop worrying about castings; they are not commercially offered in most classes.

2. DN50 to DN150 is the only genuine decision zone. Use the crossover table, then check whether a no-repair or weight clause has already decided it for you.

3. Above DN150, cast is the default for gate, globe and check, and forged is a special order justified only by a severe-service clause or an emergency.

4. Above Class 900 at DN100 and below, assume forged until a quotation proves otherwise.

5. Never accept a standard-class casting on a corrosion-sensitive duty without a radiography clause, a severity level and a repair disclosure requirement.

6. For spares and one-off replacements, price the forged-from-block option before accepting a twelve-week pattern lead time. Below roughly 150 units at DN100 it is often the cheaper and faster answer.

7. When in doubt, specify the examination rather than the process. A fully radiographed premium casting and a fully ultrasonically examined forging are both defensible; an unexamined body of either route is not.

At J-VALVES both routes are quoted with the process route, the heat-traceable certificate, the NDE standard and coverage, and the weld repair policy stated on the document set rather than left for the buyer to discover. A forged-versus-cast comparison is only meaningful when both quotations carry those four items.

12. Frequently Asked Questions

Is a forged valve body always stronger than a cast one? 

No. Both routes are qualified to the same pressure-temperature rating in ASME B16.34, and a sound premium-quality casting is rated identically to a forging. What forging removes is the class of defect that comes from solidification: shrinkage and gas porosity. What it introduces is a different class of defect, namely laps, seams and billet centreline segregation if the reduction ratio is too low. The honest statement is that a forged body has a narrower defect distribution in small sections, not that it is stronger.

At what size does a cast body become cheaper than a forged one? 

In practice, from about DN150 upward at low pressure class, and from about DN80 to DN100 upward at Class 900 and above. Below that, forging is either the only commercially available route or the cheaper one once tooling is counted. The crossover is a quantity as well as a size: at DN100 Class 300 a casting needs roughly 180 to 200 units to recover its pattern cost against a body machined from forged block, while at DN300 Class 150 the same calculation needs only about 16 units.

Can a fully radiographed casting be used instead of a forging? 

Often yes. ASME B16.34 gives a casting quality factor of 1.00, the same as a forging, to a casting upgraded to premium quality by full volumetric examination, so the wall thickness penalty disappears. Two conditions apply. The radiography must be specified with a reference standard, a severity level and a coverage rate rather than as the word 'radiographed', and the weld repair policy must still be acceptable, because full radiography does not by itself prohibit repairs unless you write that prohibition into the order.

Why did a cast valve pass its hydro test and then leak a year later? 

Because a hydrostatic shell test is not a soundness test. It proves the pressure boundary holds at test pressure, at ambient temperature, on the day of testing. A subsurface shrinkage cavity surrounded by sound metal passes without difficulty, but it sits directly under the wetted surface. Once corrosion removes a fraction of a millimetre, or thermal cycling opens a micro-crack from the void, the cavity breaches and the body weeps. In chloride or stagnant service the void also acts as a crevice that initiates pitting, which is why the failure usually appears at a flange hub or stem boss.

Is a bar stock body the same as a forged body? 

No, although API 602 permits both for small valves. A bar stock body is machined from rolled bar, so the material is wrought and clean but the grain flow is straight along the bar axis and is then cut away by the bore and the bonnet thread. A forged body is shaped by deformation, so the grain follows the contour of the body and the section is reduced in the direction that closes internal voids. Bar stock is usually 10 to 20 percent cheaper and perfectly acceptable for utility service; if your specification says forged, say so explicitly and state that bar stock is not acceptable.