​‌​​‌​

AISI 8740 / AMS 6322 / AMS 6325

AISI 8740 is a low-alloy steel, alloyed with nickel, chromium and molybdenum, that can be hardened by heat treatment. Within the alloy steel group it stands out for combining high strength with good ductility and formability.​‌​​‌​

It works at lower hardness levels than the 4140 and 4340 grades; in return its weldability and formability are better. That balance makes it the preferred grade for fasteners such as bolts, nuts and pins that work under high tensile load.

After heat treatment its microstructure consists of martensite and tempered ferrite. Its ability to hold dimensional stability after tempering is an important advantage in parts held to tight tolerances. In grades produced by vacuum arc remelting (VAR) or electroslag refining (ESR), microstructural cleanliness is of the highest order.​‌​​‌​

It is used in aerospace for landing gear parts, fasteners and propeller shafts; in automotive for valve components, gears and drive shafts; and in oil and gas for high pressure downhole equipment. It can be supplied as round bar and forgings.

What AISI 8740 Is — and Why It Is Not “4140 With Nickel”​‌​​‌​

AISI/SAE 8740 (UNS G87400 / W.Nr. 1.6546 / commonly cross-referenced in Europe as 40NiCrMo2-2, and in Italian sources as 40NiCrMo2) is a medium-carbon, triple-alloyed (Ni-Cr-Mo) THROUGH-HARDENING — quenched and tempered — steel. Its single distinguishing sentence: it is the classic aerospace fastener steel, and it earned that position not through strength but through BALANCE — adequate hardenability, good toughness, good machinability and enough ductility to roll a thread.

And two things have to be said at the outset. First: 8740 is NOT stainless. It contains about 0.50 % chromium; the passivity threshold is roughly 10.5 %. That is why aerospace 8740 bolts are almost always supplied cadmium plated — and the plating opens the most critical section on this page (hydrogen embrittlement). Second: 8740 is not the “high-carbon version” of 8620; it is a steel built for a DIFFERENT JOB. 8620 is a carburizing steel (0.18–0.23 % C) that produces a hard case; 8740 (0.38–0.43 % C) produces one uniform quenched-and-tempered structure throughout. Sharing a family number does not make them alike.​‌​​‌​

The engineering lies in a balanced distribution of three elements. 4140 raises chromium alone (0.80–1.10 %); 4340 raises nickel alone (1.65–2.00 %). 8740 keeps all three at moderate levels: 0.40–0.60 Cr · 0.40–0.70 Ni · 0.20–0.30 Mo. And that molybdenum figure deserves careful reading: 8740 leaves the 86xx family (0.15–0.25 % Mo) and joins the 87xx family on the strength of exactly that 0.20–0.30 %. Molybdenum does two jobs here: it adds hardenability and it resists reversible temper embrittlement. The second job is the subject of the embrittlement section below.

8740, 8740H, E8740 and “AMS 6322 8740” — four different purchase-order lines

8740 is a chemistry specification (ASTM A29 / A322, SAE J404). 8740H is a hardenability specification: the chemistry band widens but the mill guarantees the Jominy band. E8740 is a prefix denoting electric-arc-furnace production. “AMS 6322 8740” is none of those: it is an aerospace quality specification, and it also tightens the chemistry — notably to P ≤0.025 % and S ≤0.025 % (against 0.035 % and 0.040 % on the ASTM side). “We have 8740 bar” does not satisfy an AMS 6322 order.​‌​​‌​

Standards by Product Form

​‌​​‌​

​‌​​‌​

8740’s standards map differs from 4140’s: the centre of gravity is not ASTM, it is AMS. The reason is simple — this steel spends most of its commercial life as an aerospace fastener.

8740 as an Aerospace Fastener — the Specification Chain

This section is the key to understanding 8740. The steel spends most of its commercial life not as bar but as a bolt — and the specification for that bolt does not start with the material, it starts with the drawing.​‌​​‌​

MS21250 (12-point bolts) is a different chain, and the same number can cover different materials: A-286, Inconel 718, Ti-6Al-4V, 17-4 PH and 8740. The 8740 variant is heat treated to AMS 6322. One vendor quotes “up to 220 ksi” for 8740 in this family — that is single-source and far above 8740’s established aerospace band of 160–180 ksi; do not accept it without verifying the drawing and specification.​‌​​‌​

Why is 8740 preferred over 4340 for bolts?

The answer is not strength; it is MANUFACTURABILITY and RISK. Three reasons:
1. Thread rolling. On an aerospace bolt the thread is rolled, not cut — because a rolled thread preserves the material flow lines and multiplies fatigue life. Rolling after heat treatment requires the material to retain some ductility. 8740 in the 160–180 ksi band provides it comfortably.
2. Hydrogen embrittlement risk rises steeply with strength. Holding 8740 in the 160–180 ksi band keeps the risk manageable. Push the same bolt to 200+ ksi in 4340 and the smallest lapse in plating or baking discipline becomes a delayed fracture.
3. Repeatability and cost. 8740 reaches the 160–180 ksi band over a wide tempering window, so hitting the heat-treat target is easy. In high-volume fastener production that translates directly into scrap rate.
In short: 8740 is there not because it is the strongest steel, but because it is the steel that causes the fewest problems in the 160–180 ksi band.​‌​​‌​

Specification Gaps — What “8740” Does Not Buy You

Chemical Composition

Heat analysis and product analysis are not the same thing

The bands above are the heat (cast) analysis. For a sample taken from the finished product (product analysis), the standard allows a tolerance for production scatter. One published reference gives C 0.37–0.44 % and Mn 0.70–1.05 % for 8740 — those are product-analysis tolerances, not the heat band. Before rejecting a certificate, check which one you are looking at; buyers who do not know this distinction reject conforming material unnecessarily.​‌​​‌​

Why does phosphorus matter so much?

Because phosphorus is the PRINCIPAL AGENT of reversible temper embrittlement. Together with tin, antimony and arsenic it segregates to prior austenite grain boundaries in the 375–575 °C band and raises the ductile-to-brittle transition temperature. AMS 6322 pulling phosphorus from 0.035 % to 0.025 % is not a formality — it is a measure taken directly against this mechanism. For an 8740 part that will work at high strength, the P value on the certificate may matter more than the hardness value.​‌​​‌​

Heat Treatment — Quenching and Tempering

​‌​​‌​

​‌​​‌​

Tempering temperature versus hardness — a section to read carefully​‌​​‌​

This page will NOT give you a full tempering curve (Rm, Rp0.2, A, Z, KV at every temperature) — because it could not be independently verified from accessible sources. The published 8740 tempering tables typically belong to specimens oil quenched from 830 °C and tempered at 205 / 315 / 425 / 540 / 595 / 650 °C, in ⌀25 mm and ⌀50 mm rounds; the headings of those tables were verifiable, their numerical contents were not. Printing an unverified curve here would send you to the wrong recipe.
The individually verifiable points are:
• tempering at ≈595 °C gives ≈248 HB — described as a “balanced” working point. Single source.
• tempering in the 200–300 °C band is reported to give 50–55 HRC. Single source, but consistent with 0.38–0.43 % carbon.
• A general quenched-and-tempered band: Rm 800–1000 MPa, Rp0.2 550–800 MPa, A 10–18 %, 250–300 HB (25–35 HRC). Single source and wide; not a design value.
• The established aerospace fastener target: 1100–1240 MPa (160–180 ksi) Ftu. This is the band 8740 actually works in, and it is normally reached by high tempering.
What to do: if your target strength is fixed, heat treat a test coupon from your own section and set the tempering temperature from it. Do not use a catalogue curve as a recipe — least of all one with no section size stated.

TEMPER EMBRITTLEMENT — the Most Critical Section on This Page​‌​​‌​

THREE SEPARATE phenomena get confused here. They occur at different temperatures, work by different mechanisms and have different cures. If a datasheet lumps them under one heading, that page is a copy-paste.

8740 has molybdenum — is it immune?

No. Molybdenum REDUCES susceptibility; it does not confer immunity. That distinction is the source of the most common conflict in the published literature: one source writes that 8740 “is not subject to temper embrittlement”, while a professional body recommends that no tempering cycle below 1100 °F (≈595 °C) be used on low-alloy steels.
The honest reading: 8740’s 0.20–0.30 % Mo is higher than the 0.15–0.25 % of 8640 in the same family, and that is a real advantage. But it does not remove the mechanism — particularly when phosphorus sits at the top of its band and the part is furnace cooled slowly after tempering.
Practical rules:
1. If the part will take impact loading, do not temper in the 375–575 °C band. Stay either below it (high strength, low toughness) or above it (lower strength, high toughness).
2. If you tempered at 595 °C or above, cool RAPIDLY through 575 → 375 °C (in oil for heavy sections). A slow furnace cool puts back the embrittlement you just dissolved. That one sentence explains most of the “the heat treatment was right but the part came out brittle” cases.
3. For parts that will work at high strength, order AMS 6322 quality (P ≤0.025 %) rather than plain ASTM A322 (P ≤0.035 %).
4. A hardness test will NOT catch this damage. If you suspect it, order Charpy — and have a rapidly cooled reference specimen prepared for comparison.​‌​​‌​

Mechanical Properties

​‌​​‌​

​‌​​‌​

The table below shows the CONFLICTS between published values openly. The data in circulation for 8740 is less consistent than for 4140, and the reason is that the steel spends most of its commercial life not as bar but as a finished bolt.

The lesson from that table: the mechanical values in circulation for 8740 are published without stating condition, section or measurement method, and they do not agree. Annealed yield: 415 MPa versus 350–450 MPa. Quenched-and-tempered yield: 550 MPa versus 620 MPa. Annealed hardness: 201 HB versus 220–260 HB. Make none of them a design value or an acceptance criterion. If you are doing aerospace work, the part drawing and AMS-H-6875 bind you anyway; if you are doing industrial work, work from the certificate of your own batch.

Physical Properties​‌​​‌​

Welding — Preheat Is MANDATORY​‌​​‌​

8740 is not a “weldable” steel; it is a steel that CAN be welded but preferably is not. Carbon is 0.38–0.43 % and the reported carbon equivalent is ≈0.75 — high-risk territory in welding engineering. It is reported to exceed 54 HRC in a ⌀25 mm section on an oil quench; that means the heat-affected zone (HAZ), if it cools fast, reaches the same hardness and cracks.

When not to weld

1. On a heat-treated 8740 part in the high-strength band. Welding locally destroys the quenched-and-tempered structure; the HAZ contains both an over-hard zone and an over-tempered soft zone. If you cannot re-heat-treat afterwards, the mechanical properties of that part are no longer known.
2. On a plated (cadmium) part. Cadmium vaporises at welding temperature and the FUME IS TOXIC. It also causes liquid metal embrittlement in the weld. Never weld a plated 8740 part; strip the plating completely first.
3. On an aerospace fastener. A NAS/MS bolt is replaced, not repaired. Weld repair voids the certification.
4. In structures where post-weld heat treatment is impossible. There the right steel is 4130, not 8740 — carbon is 0.28–0.33 % and the carbon equivalent is markedly lower.​‌​​‌​

Machining

Corrosion, Cadmium and HYDROGEN EMBRITTLEMENT

AISI 8740 is NOT stainless. It contains about 0.50 % chromium; the threshold for passive behaviour is roughly 10.5 %. 8740 has about one twentieth of it, and that chromium is there for hardenability, not corrosion resistance. In practice 8740 rusts like plain carbon steel and will show surface rust within days if left unprotected.​‌​​‌​

Hence the plating — and the plating brings a new risk

The established finish for aerospace 8740 fasteners is cadmium (AMS-QQ-P-416, typically Type II Class 2). Three properties earned it that position: it is galvanically compatible with aluminium structure, its lubricity makes the torque–preload relationship repeatable, and it protects well in salt environments. As an alternative, chromium plating (AMS-QQ-C-320 Class 2) may be used on the shank.
But here the real subject begins: cadmium and chromium plating are ELECTROLYTIC. Electroplating — and the acid pickling that precedes it — charges atomic hydrogen into the steel.​‌​​‌​

Hydrogen embrittlement — 8740’s number one field failure

The mechanism: atomic hydrogen generated in the plating bath enters the steel, migrates to regions of stress concentration (thread roots, notches, prior austenite grain boundaries) and accumulates there. Under sufficient static stress, a crack nucleates and propagates at those sites. The result: the part comes out of plating sound, its hardness is right, it passes the tensile test — and then it fractures without warning under constant load, hours or days after installation. This is called delayed fracture.
Susceptibility rises with strength. The published thresholds: steels above 180 ksi (≈1240 MPa) require extended baking; above 200 ksi (≈1380 MPa) is the most susceptible zone; steels below 120 ksi (≈830 MPa) may be exempted from baking with approval.
And 8740 sits squarely in the risk band: its established aerospace fastener target is 160–180 ksi (1100–1240 MPa).​‌​​‌​

What to do when you order​‌​​‌​

1. Make the bake a purchase requirement and WRITE THE SPECIFICATION NUMBER. “Hydrogen relief will be performed” is not enough; write AMS 2759/9 and have the temperature–time–start window confirmed. Do not let an arbitrary temperature and time be applied.
2. Check the pre-plate cleaning method. Hydrogen comes not only from the plating bath but also from acid pickling and electrolytic cleaning. On high-strength parts, mechanical cleaning (abrasive blasting) should be preferred.
3. Make sure the bake temperature does not exceed your tempering temperature. 190–218 °C can be a problem for a low-tempered 8740 part. On a high-tempered (595 °C) part there is no issue.
4. Rolled threads must be formed BEFORE plating. Cold forming after plating both damages the coating and creates new stress concentrations.
5. Ask the supplier for the ASTM F519 qualification — especially if the plating line or bath chemistry has changed.

Where it FAILS​‌​​‌​

1. Unplated atmospheric exposure. It rusts.
2. Seawater, salt spray, road salt. Plating is mandatory, and protection ends the moment the plating is damaged.
3. Electroplating at high strength with the bake omitted. Delayed fracture. This is 8740’s most dangerous failure mode.
4. Sour (H₂S) service. NACE MR0175 / ISO 15156-2 typically imposes a 22 HRC maximum on carbon and low-alloy steels. 8740 in the 160–180 ksi band is roughly 34–40 HRC — far above the limit, and not suitable for sour service.
5. Anywhere a non-magnetic material is required. It is ferromagnetic.
6. Parts tempered in the 375–575 °C band that take impact. The temper embrittlement band.
7. Service temperatures approaching the tempering temperature. The part softens; 8740 is not a creep steel.
8. Welding on a plated part. Cadmium fume is toxic and it causes liquid metal embrittlement.

Frequently Asked Questions​‌​​‌​

We ordered NAS6203 bolts and the certificate says 4140. Was the wrong item shipped?

Most likely not — and this is the most misunderstood aspect of buying aerospace fasteners.
A NAS number is a PART number, not a MATERIAL number. The procurement specification for the NAS6203–NAS6210 family (NAS4002) permits three alloys: 4140 (G41400), 4340 and 8740 (G87400). Which one the manufacturer uses is their choice; what binds you is not the material name but the heat-treat band: 1100–1240 MPa (160–180 ksi) Ftu, with controls per AMS-H-6875.
So the right question is not “why did 4140 arrive?” but “were the 160–180 ksi band and the finish requirements met?” On the certificate, look for: the heat-treat band, cadmium plate to AMS-QQ-P-416 Type II Class 2, a hydrogen embrittlement relief (AMS 2759/9) statement, and the UNJF-3A thread (MIL-S-8879).
If you need a specific alloy, you have to SAY SO at drawing level. If, for reasons of magnetic permeability, low-temperature toughness or supply-chain standardisation, you require 8740 specifically, put it on the order as a separate line — the NAS number will not do it for you.​‌​​‌​

We have to choose between 8740 and 4340. Which is better?

“Better” cannot be answered without naming the target strength band. That is exactly where the distinction lies.
If you will work in the 1100–1240 MPa (160–180 ksi) band: 8740 should be preferred. Two reasons. First, hydrogen embrittlement risk rises steeply with strength, and keeping 8740 in this band keeps the risk manageable. Second, thread rolling: in this band 8740 retains enough ductility, and a rolled UNJF-3A thread raises fatigue life substantially.
If you must go above 1240 MPa (180 ksi): move to 4340. Its nickel is 1.65–2.00 % (8740 has 0.40–0.70) and it hardens far deeper, giving better fracture toughness at the same strength. But your plating and baking discipline must be flawless — at high strength 4340 is markedly more hydrogen sensitive, and there is a substantial body of published work on delayed fracture of cadmium-plated 4340.
If the section exceeds ⌀75 mm: 4340 again. 8740 will not harden to the centre at those sizes.
And a third option: if your target is the 1900 MPa band, both are wrong; look at 300M or AerMet 100.​‌​​‌​

The part was tempered at 500 °C, the hardness was correct, but it broke on impact during assembly. What happened?

Almost certainly REVERSIBLE TEMPER EMBRITTLEMENT. And the most dangerous feature of that phenomenon is exactly what you describe: IT DOES NOT CHANGE THE HARDNESS.
500 °C sits right in the middle of the embrittlement band (375–575 °C). In that band, phosphorus, tin, antimony and arsenic segregate to prior austenite grain boundaries. The result: the ductile-to-brittle transition temperature rises and fracture toughness falls. But hardness and tensile strength barely move — so the part passes your routine quality control without trouble and reveals itself only under impact. That is precisely what happened to you.
“But 8740 has molybdenum — isn’t it immune?” — No. Molybdenum (0.20–0.30 %) reduces susceptibility but does not remove it. While one publisher writes that 8740 “is not subject to temper embrittlement”, a professional body recommends that no tempering cycle below 1100 °F (≈595 °C) be used on low-alloy steels. Those two statements conflict, and the second one is on the safe side.
Check two further suspects. First: the phosphorus value. If P on the certificate sits near the top of the band (say 0.030–0.035 %), susceptibility rises markedly. AMS 6322 quality pulls P to 0.025 % — ask for it on critical work. Second: how did you cool after tempering? A slow furnace cool recreates the damage as it passes back through the band.
What to do: (1) Temper at 620–650 °C instead of 500 °C and re-evaluate the target strength accordingly. (2) After tempering, cool RAPIDLY through 575 → 375 °C (in oil for heavy sections). (3) Alternatively stay below the band — but then take care not to land in the tempered martensite embrittlement band (200–370 °C). (4) Add Charpy alongside hardness as an acceptance criterion; hardness will not catch this damage.​‌​​‌​

Our cadmium-plated 8740 bolts fractured two days after installation. Is the material defective?

Almost certainly not the material — the PROCESS. What you describe is the textbook definition of DELAYED FRACTURE from hydrogen embrittlement.
The signature: the part comes out of plating sound, its hardness is right, it passes the tensile test — and then it fractures under constant load, hours or days later, without warning. The fracture surface is typically intergranular and initiates at a thread root or a notch.
Why 8740? Because its established aerospace target band is 160–180 ksi (1100–1240 MPa), and that band is inside the susceptible zone. The published thresholds say extended baking above 180 ksi, highest susceptibility above 200 ksi, and approved exemption below 120 ksi.
Checklist — ask in this order:
1. Was the bake performed, and HOW MANY HOURS after plating did it start? AMS 2759/9 requires a start within 1–4 hours of plating. This is the most frequently violated rule. “We baked the next morning” largely explains the failure.
2. Were the temperature and time right? 190–218 °C (375–425 °F) for 2–24 hours, at the long end for high-strength parts.
3. What was used for pre-plate cleaning? Acid pickling and electrolytic cleaning also charge hydrogen. High-strength parts should be cleaned mechanically.
4. Was any cold forming or thread rolling done AFTER plating? If so, new stress concentrations were created.
5. Is the supplier’s ASTM F519 qualification current? A change of plating line or bath chemistry requires requalification.
6. Were the bolts over-torqued? Delayed fracture requires static stress, and excess preload raises the risk. Cadmium has a low friction coefficient — it produces higher preload at the same torque.
What to do: send the fractured parts for fractography (looking for the intergranular signature), extract the bake start time from the plating records, and add AMS 2759/9 and ASTM F519 to the order text. Verify the process before blaming the material.​‌​​‌​

Common datasheet errors — check these before you order

1. “8740 is a case-hardening (carburizing) steel” — WRONG. A published distributor page describes 8740 as “produced for case hardening”. 8740 is at 0.38–0.43 % carbon and is a THROUGH-HARDENING steel. The carburizing grade in the same family is 8620 (0.18–0.23 % C). The family-number similarity is what causes this error.
2. “200 °C temper → 60–62 HRC” — METALLURGICALLY IMPOSSIBLE. A published table states this. In a steel at 0.38–0.43 % carbon, as-quenched martensite peaks at roughly 57–58 HRC; after tempering at 200 °C it should sit in the 52–55 HRC band. 60–62 HRC requires carbon above about 0.6 %. Treat that entire table with suspicion.
3. Two irreconcilable annealed hardness values circulate. One source gives 201 HB, another 220–260 HB. They cannot describe the same condition.
4. Two different quenched-and-tempered yields are published. For the same Rm (930–938 MPa) and the same hardness (269 HB), one publisher gives 550 MPa and another 620 MPa — a 13 % gap. Do not make it an acceptance criterion.
5. Thermal conductivity and expansion conflict. Conductivity is quoted as 39 and as 46.6 W/m·K; expansion as ≈13 ×10⁻⁶ /K and as 6.6–11.9 ×10⁻⁶ /°C — two irreconcilable sets. Do not trust a single published number in a shrink-fit calculation.
6. A single “8740 strength” number is published. The same steel can sit anywhere between 695 MPa (annealed) and 1240 MPa (aerospace band). Any mechanical value quoted without a condition and a section is meaningless.
7. Tempering curves are given with no section size. The published 8740 tables typically belong to ⌀25 mm and ⌀50 mm specimens. The same values cannot be reached at the centre of a ⌀100 mm bar.
8. “8740 is not subject to temper embrittlement” — CONFLICTING. Molybdenum (0.20–0.30 %) reduces susceptibility; it does not confer immunity. A professional body recommends no tempering cycle below 595 °C on low-alloy steels.
9. Two different embrittlement phenomena get confused. Tempered martensite embrittlement occurs at 200–370 °C and is IRREVERSIBLE; temper embrittlement occurs at 375–575 °C and is REVERSIBLE. Different mechanism, different cure.
10. If an 8740 page mentions “475 °C embrittlement”, that page is a copy-paste. That phenomenon concerns steels at 15 % chromium and above; 8740 has 0.40–0.60 %.
11. The COOLING RATE after tempering is never mentioned. Rapid cooling through 575 → 375 °C prevents the embrittlement dissolved at high temperature from returning. That one step explains many “the heat treatment was right but the part came out brittle” cases.
12. “AMS 6322 = 8740” — INCOMPLETE. AMS 6322 tightens the chemistry to P ≤0.025 % and S ≤0.025 % (against 0.035 % and 0.040 % on the ASTM side). The ASTM A322 bar in your rack may not satisfy an AMS 6322 order.
13. AMS numbers are used interchangeably. 6322 bars/forgings/rings, 6323 mechanical tubing, 6325 and 6327 bars and forgings (different combinations), 6358 sheet/strip/plate, 7452 bolts and screws, 7456 studs, 7496 flash-welded rings. They are not interchangeable.
14. “NAS6203 = 8740” — WRONG. Under NAS4002, 4140, 4340 or 8740 may be used. The NAS number does not fix the material; it fixes the heat-treat band (160–180 ksi).
15. Hydrogen embrittlement is not mentioned, or is waved away as “it gets baked”. The real requirements are numerical: 190–218 °C, 2–24 hours, start within 1–4 hours of plating, AMS 2759/9, verification by ASTM F519. A “bake-out” commitment with no specification number is not a commitment.
16. How critical the baking window is goes unstated. Delay reduces the effectiveness of the bake. “We baked the next day” is a nonconformance.
17. The TORQUE–PRELOAD effect of cadmium is ignored. Cadmium is slippery and produces higher preload at the same torque. If the finish changes, torque values must be recalculated.
18. Welding on a plated part is suggested. Cadmium fume is toxic and it causes liquid metal embrittlement. The plating must be stripped completely first.
19. Cutting speeds are quoted HIGHER than 8620’s. 8740 carries twice the carbon and is harder; if the published turning speed is 305–370 m/min while 8620’s is 245–335 m/min, the two sets are internally inconsistent. Read them as ceilings.
20. The European equivalent is presented as settled. The 1.6546 ↔ 40NiCrMo2-2 cross-reference is common but could not be independently verified. Compare the chemistry line by line before issuing a dual certificate.
21. Product analysis and heat analysis get confused. The heat band is C 0.38–0.43 % · Mn 0.75–1.00 %; the published product-analysis tolerance is C 0.37–0.44 % · Mn 0.70–1.05 %. Do not reject a conforming certificate over this.
22. The threading method is left out of the specification. In aerospace the thread is rolled, and rolling after heat treatment gives the highest fatigue life. A cut thread on the same bolt gives markedly lower life.
23. “8740 behaves almost like stainless” — WRONG. The passivity threshold is 10.5 % Cr; 8740 has 0.40–0.60 %. It rusts and it depends on plating.
24. NACE compliance is claimed. The limit is 22 HRC; 8740 in the 160–180 ksi band is roughly 34–40 HRC. It is not suitable for sour service.​‌​​‌​

​‌​​‌​

​‌​​‌​