Nov 13, 2026 Leave a message

Why AMS 5536, ASTM B435, ASME SB 435 and AMS 5754 Collide on the Same Purchase Order

An industrial furnace builder in northern Italy sends a drawing for a radiant-tube support plate. The material block says "Hastelloy X, AMS 5536, 8 mm." The mill quotes it. The plate arrives with a certificate headed ASTM B435. The customer's quality department rejects it, not because the metal is wrong - the metal is exactly right - but because the document does not match the drawing. Six weeks are lost over a specification family that describes the same alloy in the same condition.

 

That is not an unusual case. It is the standard failure mode for this grade. Alloy X (UNS N06002) is one of the few nickel alloys that is simultaneously an aerospace sheet material and an industrial furnace plate material, and the two worlds wrote different paperwork for it. This article maps the four designations that appear on purchase orders, shows where each one actually governs, and lists the six places where an order goes wrong even when everybody involved is competent.

 

The alloy itself is straightforward: roughly 22% chromium for oxidation resistance, 18% iron for cost and expansion balance, 8–10% molybdenum plus tungsten and cobalt for solid-solution strength, carbon 0.05–0.15% for grain-boundary carbide strengthening. Density 8.22 g/cm³, melting range 1,260–1,355 °C, fully austenitic and paramagnetic. Nothing exotic. The difficulty lives entirely in the documents.

 

The map: what each document is allowed to cover

Designation Product forms covered Condition Typical buyer
AMS 5536 (rev R, 2024) Sheet, strip, plate Solution heat treated Aerospace prime, gas turbine OEM, defence
AMS 5754 Bar, forgings, rings Solution heat treated Same customers, machined parts
ASTM B435 / ASME SB-435 Plate, sheet, strip Solution heat treated Furnace builders, chemical plant, EPC
ASTM B572 / B619 / B622 / B626 Rod, welded pipe, seamless tube, welded tube As specified Piping and tubular packages
ASTM B366 Fittings As specified Piping spools
AMS 5798 / 5799 Welding wire / covered electrodes Filler metal Fabricators consumables

 

Note what the table implies: AMS 5536 and AMS 5754 split the wrought product range at the product form, not at the alloy. Same chemistry limits, same minimum mechanical properties, same heat-treatment condition - different document because one covers flat product and the other covers long product. A buyer who writes "AMS 5536" on a bar order, or "AMS 5754" on a plate order, has written a document that the mill cannot certify against, even though the mill can supply the metal.

 

Mistake one: ordering a plate against the bar specification, or the reverse

This is the most common and the cheapest to avoid. The two AMS documents are siblings, not substitutes. If the drawing is a formed and welded sheet structure - a combustor liner, a transition duct, a muffle - the governing document is AMS 5536. If the drawing is a machined flange, a shaft, a ring, a forged manifold, the governing document is AMS 5754.

 

The practical test is the first operation in the route card. If the first operation is shear, laser, waterjet or press forming, it is flat product and you want 5536. If the first operation is saw or turn, it is long product and you want 5754. A purchase order that names the wrong one is not a paperwork slip that a certificate can fix; the mill will either decline to certify or will issue a certificate with a scope statement that does not cover the form delivered, and the receiving inspection will catch it.

 

Mistake two: assuming AMS and ASTM differ in the numbers

They do not, and this is where buyers over-correct. Both families land on the same room-temperature floors for solution-treated flat product:

Property Requirement MPa equivalent
Tensile strength, min 100 ksi 690
Yield strength, 0.2% offset, min 45 ksi 310
Elongation in 2 in, min 35% -

 

The difference is in how those numbers are proved. AMS 5536R, issued in 2024, tightened exactly this layer: it added strain-rate requirements to the tensile test, clarified the test procedure, revised finish requirements, updated composition testing and reporting, and restated its prohibition on exceptions. It also invokes AMS 2262 for thickness tolerances, AMS 2269 for chemical check analysis limits, AMS 2283 for composition test methods, AMS 2371 for sampling, AMS 2750 for pyrometry and AMS 2807 for identification marking - a full quality chain that an industrial B435 order does not carry.

 

So the honest position for a buyer is: the metal is interchangeable, the evidence package is not. If the end user is a turbine OEM or a defence programme, AMS 5536 is mandatory because the customer's own quality system requires that chain. If the end user is an industrial furnace or a chemical plant, B435 is sufficient, and paying for the aerospace chain buys nothing except a longer lead time and a higher price. Writing "AMS 5536" onto an industrial order because it sounds stricter is a real cost with no returned value.

 

Mistake three: ignoring the 4.76 mm line

Both plate documents cover sheet and plate in one text, but the tolerance regime changes at a specific thickness: sheet is up to and including 0.1874 in (4.76 mm), plate is 0.1875 in (4.76 mm) and above. The same boundary logic runs through the nickel plate standards generally.

 

Three things move when you cross that line: thickness tolerance, camber and flatness allowance, and edge condition. A 4.5 mm and a 5.0 mm item are not the same product, even when both are described as Alloy X plate by a sales office. If a drawing calls for 4.76 mm exactly - and plenty of legacy aerospace drawings do, because it is 3/16 in - confirm with the mill whether the item will be rolled as sheet or as plate before the price is fixed. The step at that boundary is visible in the quotation.

 

hastelloy x plate

Mistake four: treating "solution heat treated" as a phrase rather than a temperature

Every document in the table says the product is supplied solution heat treated. None of them tells you the number, and that is where plate gets ruined quietly. Producer practice for this alloy sits in a window around 1,150–1,205 °C (2,100–2,200 °F), most commonly 1,177 °C (2,150 °F), followed by rapid cooling - water quench, or rapid air for thin sections, with bright-annealed product cooled in hydrogen.

 

Anneal below that window and you do not get a softer plate. You get a plate with precipitated secondary phases: room-temperature strength rises slightly, ductility falls, and impact toughness drops. It will pass a tensile test. It will fail a bend test. In service it may fail on the first thermal cycle rather than the ten-thousandth.

 

This is why the purchase order should carry the heat-treatment line explicitly: solution anneal at 1,175–1,205 °C, hold to temperature through section, rapid cool, cooling method reported on the certificate. A certificate that says only "solution annealed" is a statement of intent, not evidence. The single most useful sentence a buyer can add to the order is the one that asks for the anneal temperature and the cooling method to be printed on the mill certificate.

Mistake five: buying an oxidation alloy for a corrosion duty

This grade is specified for heat, not for chemistry. Its 22% chromium produces a tenacious Cr2O3 scale that resists spalling to about 1,204 °C (2,204 °F) in oxidizing atmospheres. That is the design intent. It is not a wet-corrosion alloy.

 

Concretely: no seawater immersion, no chloride brines, no reducing acid service. A 9% molybdenum nickel alloy will survive many acidic streams, and its corrosion resistance is closer to high-nickel grades than a datasheet reader might expect, but it has nothing like the molybdenum level of C-276 and no niobium to match 625. A plant that specifies Alloy X for a wet chloride stream because the temperature is also high has solved the temperature problem and created a pitting problem.

 

Where the alloy genuinely shines beyond plain oxidation is in carburizing and nitriding atmospheres, which is a furnace problem rather than a chemistry problem. In petroleum coke at 1,038 °C, a 309 stainless specimen was fully penetrated within 100 hours while Alloy X showed no measurable carburization depth. In a hydrogen/nitrogen/ammonia atmosphere at 593 °C and 15,000 psi, apparent nitride layer was 0.01 mm against 0.08 mm for 310 stainless, with no intergranular attack and no bend failure. Those two data sets are the reason ethylene furnaces and heat-treat shops specify it.

 

Mistake six: designing for the strength and forgetting the toughness

This alloy is solid-solution strengthened and cannot be age hardened. Between roughly 650 °C and 982 °C, however, it does precipitate secondary phases over time, and the consequences show up in impact energy rather than in tensile strength:

Condition Charpy V-notch, room temperature
Solution treated, as supplied 54 ft-lb / 73 J
After 168 h at 816 °C (1,500 °F) 9 ft-lb / 12 J
After 500 h at 871 °C (1,600 °F) 9 ft-lb / 12 J
After 500 h at 982 °C (1,800 °F) 20 ft-lb / 27 J

 

A plate that leaves the mill at 73 J and spends a year at 816 °C comes back at roughly 12 J. It has not failed. It has not changed shape. It has simply stopped being a ductile material. The consequences are all in handling and shutdown: cold straightening, flame cutting, chipping slag, dropping a fixture, or a fast cold start can now crack a component that would have shrugged the same treatment when new.

 

Design guidance that follows: avoid sharp stress concentrations in components that will see long service in that band; do not plan post-service cold straightening; specify the handling temperature in the maintenance procedure; and if the part must also survive wet shutdowns or washdowns, treat the toughness loss as a corrosion-risk multiplier, because a sensitized boundary plus a brittle matrix is a worse combination than either alone.

 

The weld is where the documents stop and the physics starts

Weldability is genuinely good - better than 718 or Waspaloy - and the reason is structural: with no ageing reaction there is no strain-age cracking, which is why complex multi-pass sheet assemblies are built from this alloy rather than from a precipitation-hardening grade.

Item Practice Consequence if ignored
Filler metal ERNiCrMo-2 (AMS 5798 wire); ERNiCrMo-3 acceptable for many joints Mismatched chemistry drives dilution cracking and lower creep life
Preheat Not required -
Interpass temperature Below 93 °C (200 °F) Microfissuring in the heat-affected zone
Post-weld heat treatment Usually not required; 871 °C for 1 h on large restrained fabrications; full solution anneal for severe service Residual restraint plus service cycles equals cracking
Submerged arc welding Not recommended High heat input and slow cooling raise restraint and promote cracking
Groove preparation Larger root clearance than a stainless procedure Incomplete penetration, because conductivity is about a fifth of carbon steel at 9.1 W/(m·K)

 

The thermal conductivity number deserves its own sentence, because it is the one that transfers badly from stainless steel procedures. At 9.1 W/(m·K) at room temperature, heat does not leave the weld pool quickly. A procedure taken from a 310S job will over-heat this alloy, and the damage appears as heat-affected-zone microfissuring that no radiograph will find until it becomes a leak. Weld procedure qualification, not just welder qualification, is the control.

 

Where it sits against the alloys it usually replaces

Grade Oxidizing ceiling Strength above 900 °C Wet chloride Fabricability Relative plate cost
310S / 310H ~1,050–1,150 °C Low Aqueous only, moderate Excellent Lowest of the group
Incoloy 800H / 800HT ~1,100 °C Moderate Good in many streams Good Mid-low
Inconel 625 ~980 °C Lower above 950 °C Excellent Excellent Highest
Hastelloy X (N06002) ~1,204 °C intermittent Highest in this band Not intended Good, weldable without ageing Mid

 

The substitution logic that shows up in real projects is simpler than the table suggests. 310S wins when cost dominates and the part is a simple heat-resistant shape. 800H is downgraded in when duty is 900–1,000 °C with modest load. 625 is pulled in when the same part also meets wet chlorides or needs room-temperature strength. Alloy X is specified when temperature is high and the part is a formed, welded sheet structure carrying mechanical load - the combustion-zone component that must hold shape through thousands of cycles.

 

What the elevated-temperature numbers actually let you design

The tensile table everybody quotes - 340 MPa yield at room temperature falling to 45 MPa at 1,093 °C - is almost useless above 800 °C, because a furnace part is not loaded once. It is loaded continuously for years. The number that governs is creep, and creep numbers are not in the tensile table.

 

Approximate readings from published producer creep curves for solution-treated plate, stated as initial stress to produce 1% creep in 1,000 hours:

Temperature Stress for 1% creep in 1,000 h Practical meaning
649 °C (1,200 °F) ~145 MPa (21 ksi) Design still resembles a strength calculation
760 °C (1,400 °F) ~66 MPa (9.5 ksi) Thickness is now creep-governed, not yield-governed
871 °C (1,600 °F) ~19 MPa (2.7 ksi) Section thickness grows sharply for modest load
982 °C (1,800 °F) ~1.4 MPa (0.2 ksi, extrapolated) Only lightly loaded shapes survive; support by design, not by section

 

Treat those as curve readings, not code allowables - use the applicable ASME Section II Part D values for binding design. The shape of the fall is the point: between 760 °C and 871 °C the usable stress drops by a factor of about three and a half, and between 871 °C and 982 °C by more than a factor of ten. This is why furnace engineers stop talking about alloy strength above 900 °C and start talking about geometry: span, hanger spacing, sag allowance, and how the part is supported rather than how thick it is.

 

It also explains the most common specification error in this temperature range: taking the room-temperature 310 MPa yield, applying a safety factor, and arriving at a wall thickness that will sag within a year. The material is not failing to meet its certificate. The calculation used the wrong curve.

 

Forming: hot, cold, and the trap between them

This alloy forms well, which is one of its commercial advantages over 718 and Waspaloy, but the forming rules are not symmetric between hot and cold work.

 

Hot forming requires the entire piece - not the surface - to reach about 1,177 °C (2,150 °F) and soak through before deformation begins, and the work must finish above roughly 950 °C. Parts heated only on the outside crack at the outer radius while the core is still cold and stiff. After hot working, the part must be re-solution-treated and rapidly cooled, because hot work leaves a microstructure that will not deliver the specified properties.

 

Cold forming is practical and the work-hardening rate is close to austenitic stainless steel, so existing press tooling generally transfers without redesign. The trap is the intermediate anneal: a fabricator who needs two forming stages and anneals between them at 1,050 °C because that is what the furnace is set to will precipitate secondary phases and destroy ductility in the blank. Every intermediate anneal on this alloy is a solution anneal or it is damage.

 

Machining runs at 40–80 surface feet per minute with carbide, positive-rake inserts, and flood coolant. Dry machining work-hardens the surface layer within a few passes and the next pass fails. This alloy machines more easily than 718, roughly like C-276, and considerably less easily than 304.

 

A failure review worth reading before the next order

The following is a composite of field cases that repeat with enough regularity to be worth writing down. A gas turbine combustor liner, 3 mm sheet, cyclic duty peaking near 1,050 °C, cracked through the heat-affected zone at a weld toe after roughly two years of operation. The material certificate was correct. The chemistry was correct. The tensile results were above the minimum. Three separate causes were found, and none of them was visible on the paperwork.

 

Cause one: no post-weld solution anneal. The liner was fabricated as-welded because the procedure said PWHT was not required - which is true for many duties, but not for a part that will run for years in the carbide-precipitation band. The heat-affected zone entered service already sensitized, and repeated thermal cycling then opened it.

 

Cause two: interpass temperature was never measured. The shop used the stainless procedure's feel for heat input. With thermal conductivity at roughly a fifth of carbon steel, the sheet stayed hot far longer than the welder expected, and interpass temperature ran well above the 93 °C limit. Heat-affected-zone microfissuring was present from the first pass and only grew.

 

Cause three: a fast cold start. The component had been in service long enough for room-temperature Charpy energy to fall toward the 12 J range. A cold start after a weekend shutdown introduced the thermal shock that a ductile plate would have absorbed and a brittle one would not.

Three independent mistakes, all made by competent people, none of them detectable in a receiving inspection. That is the nature of a high-temperature alloy: the certificate protects the delivery, and only the process protects the service life.

 

Why two quotes for the same plate can differ by forty percent

Buyers comparing offers on this grade regularly see spreads that look like somebody is cheating. Usually nobody is. The spread is the sum of five legitimate variables:

Variable Effect on price How to compare fairly
Certification level AMS 5536 with full aerospace chain costs more than B435 for identical metal Ask which document the certificate will cite
Thickness Heavy plate carries rolling and heat-treatment penalties; the band widens above 25 mm Quote against the same nominal and same tolerance basis
Alloy surcharge timing Nickel, molybdenum and cobalt moves are passed through, sometimes with a lag Ask for the surcharge basis date, not just the total
Stock versus mill roll Material pulled from European or US stock prices against that market, not Asian mill output Ask origin and whether the item is ex-stock
Cutting and testing Waterjet, PMI, UT and third-party witness add real cost per piece Compare landed price with the same scope of supply

 

The one line that most often explains a big gap is the first: an offer quoting B435 metal against a drawing that calls for AMS 5536 is not a cheaper offer, it is a different product, and it will be rejected at receiving inspection. The second most common is stock origin - a plate sitting in a European service centre is priced against European replacement cost, which has nothing to do with Chinese mill output that week.

 

A purchase-order template that survives receiving inspection

Line Wording What it prevents
1. Standard and form "AMS 5536R, plate, solution heat treated" or "ASTM B435 / ASME SB-435, plate" Certificate scope mismatch at goods-in
2. Alloy identity "UNS N06002 (Alloy X / HX, W.Nr. 2.4665)" Substitution with a corrosion-grade Hastelloy
3. Thickness basis Nominal or minimum, plus the applicable AMS 2262 class Weight, price and fit all move with this one word
4. Heat treatment Anneal 1,175–1,205 °C, hold through section, rapid cool, method reported A low-temperature anneal that passes tensile and fails in service
5. Grain size ASTM 3–6 where creep life is designed in; omit otherwise Paying for a clause nobody checks
6. Testing Tensile, product analysis, UT per ASME VIII where pressure service applies, PMI on 100% for critical duty Mixed heats on a grade that looks like every other nickel plate
7. Surface Hot rolled and pickled, or bright annealed, finish class stated Oxide carry-over into the furnace as a life penalty
8. Edge Sheared, saw cut or waterjet, with heat-affected allowance defined Plasma-cut edge used as a weld-prep edge

 

Indicative 2026 pricing ex-stock Asia runs about USD 35–85 per kilogram for plate, higher at heavy thickness and aerospace certification, lower for standard industrial plate. Nickel, cobalt and molybdenum movements, plus low-volume rolling schedules, move this grade more than any published list.

 

FAQ

Is Hastelloy X the same as UNS N06002?

Yes. Hastelloy X is the Haynes trade name, Alloy X or HX is the generic name, UNS N06002 is the unified number, and W.Nr. 2.4665 with NiCr22Fe18Mo are the European designations. One alloy, four naming systems.

 

Can AMS 5536 and ASTM B435 be substituted for each other?

The metal is interchangeable; the evidence package is not. AMS 5536 carries aerospace sampling, strain-rate control in tensile testing, pyrometry and identification requirements that B435 does not. Aerospace and defence customers require the AMS chain. Industrial furnace and chemical customers do not need it and should not pay for it.

 

What is the maximum service temperature of this plate?

Oxidation resistance is quoted to about 1,204 °C (2,204 °F) and useful strength to roughly 816 °C. Above 900 °C, design must be governed by creep rupture and the applicable allowable stress table, because tensile data no longer describes sustained-load behaviour. Continuous service around 1,100 °C is practical; above that, scale growth accelerates and component life falls quickly.

 

Does it need post-weld heat treatment?

Generally not. No preheat, interpass below 93 °C, and no ageing cycle. Large or highly restrained fabrications are usually stress relieved at 871 °C for one hour, and severe service justifies a full solution anneal at about 1,175 °C followed by rapid cooling.

 

Why does a used plate crack when a new one would not?

Long service in the 650–870 °C band precipitates grain-boundary phases and drops room-temperature Charpy energy from about 73 J to about 12 J. The plate is still strong and still the right shape, but it is no longer ductile. Cold straightening, chipping, or a fast cold start can crack it.

 

Is it suitable for seawater or chloride service?

No. This is an oxidation and high-temperature structural alloy. For wet chlorides, seawater and reducing acids, C-276 or 625 are the correct choices; specifying Alloy X there solves the temperature problem and creates a pitting problem.

 

Which filler metal should the fabricator use?

ERNiCrMo-2 to AWS A5.14, supplied as AMS 5798 wire for GTAW and GMAW. ERNiCrMo-3 is acceptable for many joints. Submerged arc welding is not recommended because of heat input and slow cooling.

 

Ordering this plate

If the drawing already names the standard and the form, the remaining work is the heat-treatment line and the thickness basis - those two decide whether the certificate you receive actually describes the plate you need.

 

Huitong supplies Hastelloy X (UNS N06002) plate and sheet to AMS 5536 and ASTM B435 / ASME SB-435, 3–60 mm, cut to size, with mill test certificates, third-party inspection and PMI on request.

 

Send the drawing, the design temperature and the atmosphere the part will see to market@htpipe.com, or message us on WhatsApp at +86-19339900201. We will confirm which of the four documents your order actually needs before the plate is rolled - and if B435 is enough, we will say so rather than sell you an aerospace certificate you cannot use.

Contact now

 

 

Send Inquiry

whatsapp

Phone

E-mail

Inquiry