​‌​​‌​

EN AW 7075 / AMS 4078 / AMS 4124

​‌​​‌​

AMS 4078 / AlZn5.5MgCu / 7075 / T7351

EN AW 7075 T651 is used in many fields, aerospace above all. Because its resistance to stress corrosion cracking is insufficient in large diameters and heavy thicknesses in particular, 7075 T7351 is preferred. Where the material is to be used as large single-piece aluminium plate, selecting the appropriate AMS and ASTM standards means that 7075 T7351 provides high resistance to stress corrosion while also giving machinability to near-perfect close tolerances without distortion problems.​‌​​‌​

Machinability: While EN AW 7075 provides high strength and temperature capability, it is generally a harder alloy and more difficult to machine. There are therefore some factors to observe during machining and forming.

Turning and milling: Cutting speed — EN AW 7075 aluminium alloy can generally be machined at medium cutting speeds. Higher speeds can cause overheating, which can adversely affect the machining process. Cutting tools — it can be machined with carbide inserts or hardened steel tooling, and tool life varies with the speed and the cutting fluid used. Cooling — using cutting fluid at high machining speeds prevents the material from overheating and the quality of the work from deteriorating.​‌​​‌​

Weldability: EN AW 7075 can be welded by the TIG and MIG methods, but care is required because high thermal stress can arise during welding. Surface cleanliness is important against oxidation problems during welding.

Heat treatment: EN AW 7075 can generally be given the T5 or T6 temper, which is used to raise the hardness and strength of the material. After heat treatment the material gains higher strength and hardness, although machinability can fall somewhat.​‌​​‌​

What EN AW-7075 Is — and Why the T6 / T73 Difference Is Everything​‌​​‌​

EN AW-7075 (chemical symbol EN AW-AlZn5.5MgCu / W.Nr. 3.4365 / AA 7075 / UNS A97075 / old DIN name AlZnMgCu1.5) is the standard member of aluminium’s highest-strength commercial family. Its nominal composition is 5.1-6.1 % Zn, 2.1-2.9 % Mg, 1.2-2.0 % Cu and 0.18-0.28 % Cr. In the T651 temper the measured typical tensile strength is 572 N/mm² with a proof stress of 503 N/mm² — in the band of structural steel at one third of steel’s density.

But those numbers are not the real subject of this page. The real subject is this: 7075 is one of the rare engineering materials that is deliberately weakened because it is susceptible to stress corrosion cracking (SCC). The T73 temper was invented to eliminate a failure mode by knowingly giving up about 10-15 % of the strength. That is the single most important thing to know about 7075, and most datasheets either never say it or mix T6 and T73 in the same row.​‌​​‌​

Why overageing stops stress corrosion cracking​‌​​‌​

The strength of 7075 comes from finely dispersed η′ and η (MgZn₂) precipitates in the matrix. At peak ageing (T6) these precipitates are very fine and sit as a continuous string along the grain boundaries. That continuous string forms an anodic crack path in a chloride environment under tensile stress; hydrogen transport to the grain boundary is added on top, and the result is intergranular stress corrosion cracking.

What T73 does is break up that continuous string. In the two-stage overageing cycle the grain-boundary precipitates coarsen and separate, and at the same time the precipitate-free zone (PFZ) on either side of the boundary widens. The result: the anodic path is no longer continuous and the crack cannot advance. Because the matrix precipitates coarsen too, strength falls — two faces of one coin, and you cannot have one without giving up the other.​‌​​‌​

The measurable evidence is electrical conductivity. As the precipitates coarsen, less alloying element remains in solid solution and conductivity rises: 32-35 % IACS in T6, 38-42 % IACS in T73. This is why correct overageing in 7075 is verified by measuring hardness and conductivity together. Hardness alone is not enough — an under-overaged part can be as hard as T6 and as crack-prone as T6.

Standards by Product Form

​‌​​‌​

​‌​​‌​

7075 lives in two separate specification worlds — European (EN) and American (ASTM/AMS/QQ-A) — and the two give different numbers. The aerospace side works mainly through AMS; the commercial side uses the ASTM B series; Europe uses EN 485 / EN 755 / EN 754 / EN 586. The same part can have three different sets of minima.

Product Forms With NO Standard

Chemical Composition

CONFLICT NOTICE — composition band. A widely used encyclopaedia entry gives 5.6-6.1 % Zn, 2.1-2.5 % Mg, 1.2-1.6 % Cu for 7075. Those are NOT specification limits. Producer datasheets and EN 573-3 give 5.1-6.1 % Zn, 2.1-2.9 % Mg, 1.2-2.0 % Cu on both the European and the American side — a wider band. The narrow band is most likely a TYPICAL composition range, or it belongs to a more tightly controlled variant such as 7175. Use the wide band when auditing a certificate; do not reject a heat by mistaking the narrow band for a specification limit.

Mechanical Properties​‌​​‌​

In 7075 the gap between “minimum” and “typical” is more dangerous than in other alloys, because two specification worlds circulate at the same time. The three are given separately below: American typical, American minimum and European (EN 485-2) minimum.​‌​​‌​

MIND THE GAP BETWEEN THE TWO WORLDS. On the European side, T651 plate carries minima of Rm ≥540 / Rp0.2 ≥460-470, while American producer tables give 572 / 503 as typical and 531 / 455 as minimum. These three sets do not contradict each other — they are different things: one is the European standard minimum, one the American specification minimum, one a measured typical. Document which source you calculated from, and when auditing a certificate check which specification the material was ordered to.​‌​​‌​

Physical Properties

Heat Treatment — and Why There Is No Tolerance At All

​‌​​‌​

Solution treatment​‌​​‌​

The published solution treatment is 466 °C ± 6 °C (870 °F ± 10 °F). The time runs from 30 minutes to 2 hours with section thickness; a typical approach is about 30 minutes plus 15-30 minutes per additional inch of thickness. Furnace uniformity of ± 10 °F (± 6 °C) is required, tightening to ± 5 °F on aerospace work.

Why is the tolerance this tight? Because the non-equilibrium eutectic melting point of 7075 sits around 477 °C — only eleven degrees above the solution treatment temperature. The 480 °C solidus given by a European mill sheet corroborates this. If one spot in the furnace overshoots the target by ten degrees, local melting begins at the grain boundaries (incipient melting), and THAT DAMAGE IS IRREVERSIBLE: even if the part is aged correctly afterwards, ductility, toughness and fatigue life are permanently lost, and usually nothing is visible to the eye. In 7075 heat treatment, furnace survey work (of the AMS 2750 kind) is not a formality but a safety requirement.​‌​​‌​

The quench — the most critical step

7075 is among the most quench-sensitive aluminium alloys there are. The published practice:
· Medium: cold water below 29 °C (85 °F) with vigorous agitation.
· Transfer time: a limit of about 15 seconds from furnace to quench tank for heavy sections. Exceed it and the part begins precipitating in air and cannot reach peak strength.
· Critical cooling rate: about 278 °C per second (500 °F/s) through the critical range of 399-288 °C (750-550 °F).
· Polymer quenchants: usable to reduce distortion but they carry a 5-10 % strength penalty against cold water. On a tight-tolerance part that is an engineering choice and it should appear on the certificate.​‌​​‌​

The quench has two side effects and both are expensive. The first is distortion. The second is residual stress: a rapidly quenched heavy plate carries compression at the surface and tension in the core. The “51” suffix in the T651 and T7351 tempers is exactly the answer to that — controlled stretching after the quench erases the residual stress. If you will machine precision parts from heavy 7075 plate, a stretched temper (T651 / T7351) is mandatory, not a preference.

Ageing — the T6 and T73 cycles​‌​​‌​

Can 7075 be re-heat-treated​‌​​‌​

Yes, but there is a price every time. Taking a T73 part back to T6 requires re-solution treatment, re-quenching and T6 ageing. That carries three risks: (1) renewed distortion and renewed residual stress — and stretching is no longer possible, because the part has been machined; (2) grain coarsening; (3) the risk of incipient melting at 466 °C. The other direction (T6 → T73) is easier: simply applying the overageing cycle is often enough. Practical rule: buy the temper you want; do not decide the temper afterwards.

Welding — 7075 IS NOT WELDABLE​‌​​‌​

No hedging will be used in this section. 7075 is not joined by fusion welding. One producer’s weldability rating is D (not recommended); on a European mill’s 1-6 scale, gas, TIG and MIG fusion welding all score 6 — the worst possible mark. On the same scale, resistance spot welding scores 2 — so the problem is not heat, it is the solidification of the molten pool.

Why: hot (solidification) cracking​‌​​‌​

Three mechanisms overlap:
1. A wide freezing range. According to a European mill sheet the melting range of 7075 is 480-640 °C — a band 160 degrees wide. In a weld pool solidifying across so wide a band, a thin liquid film remains at the grain boundaries in the final stage of solidification. At the same time the weld generates shrinkage stress and that liquid film tears. This is solidification (hot) cracking.
2. The composition sits at the peak of cracking sensitivity. Within the Al-Zn-Mg-Cu system the Cu and Mg levels of 7075 are in the range of maximum hot-cracking susceptibility. One source puts it plainly: “the microstructure in the weld zone of 7075 aluminium is highly susceptible to hot cracking.“
3. Liquation cracking in the HAZ. Immediately beside the fusion line, low-melting phases at the parent metal’s grain boundaries partially melt and tear under shrinkage stress. This cannot be solved by changing the filler — the problem is in the parent metal.

Why changing the filler does not solve it​‌​​‌​

Some sources cite ER5356 wire (4.5-5.5 % Mg) for TIG. That is an attempt to mitigate hot cracking, not a structural solution. For three reasons:
· HAZ liquation cracking cannot be prevented with filler metal — the problem is the parent metal itself.
· The weld metal and HAZ are re-solutionised and cool without control; the result is effectively a naturally aged, T6-like region that is WIDE OPEN to SCC. So even if the weld succeeds, you have placed exactly the failure mode you were avoiding right beside the bead.
· Hydrogen porosity. Liquid aluminium dissolves large amounts of hydrogen and on rapid solidification the hydrogen has no time to escape; the result is a porous bead.
And there are weldable 7xxx alloys: the COPPER-FREE Al-Zn-Mg alloys such as 7005, 7020 and 7039 can be fusion welded and naturally re-age by themselves in the HAZ. What makes 7075 weldable is removing the copper — and at that point what you have is no longer 7075.

How 7075 is joined​‌​​‌​

Machining​‌​​‌​

Here a datasheet conflict has to be stated openly. One producer’s summary table shows a machinability rating of D (poor) for 7075. That contradicts general industry experience and the rating given by a European mill, which gives 2 (good) for machinability in T651. The most likely explanation is that the D rating belongs to the ANNEALED (O) temper: annealed 7075 is soft and gummy and does not break chips. Aged 7075 (T6/T651/T73), on the other hand, is one of the best-machining aluminium alloys, which is why it is chosen for moulds, cutting tools and high-volume precision parts.

Corrosion — WHERE IT FAILS

​‌​​‌​

The corrosion record of 7075 is the worst of the three alloys covered on this site, and that needs to be said honestly. One producer’s general corrosion rating is C; on a European mill’s 1-5 scale, seawater and weather resistance are both rated 4-5 (poor to very poor). The reason is a single element: 1.2-2.0 % copper.​‌​​‌​

WHERE IT FAILS — 1: STRESS CORROSION CRACKING (the main subject of 7075)

This is not a corrosion problem but a FRACTURE problem, and it gives no warning. It occurs when three conditions are met simultaneously: a susceptible microstructure (T6) + tensile stress + a moist or chloride environment. The result is a crack that advances along the grain boundaries and separates the part without macroscopic plastic deformation.​‌​​‌​

The numerical classification — NASA MSFC-STD-3029:
· Table I (high resistance): “no stress corrosion failures occur on specimens stressed to 75 percent of the yield strength within 30 days.” 7075-T73, T7351, T76 and T7651 are in this table.
· Table II (moderate resistance): the same statement at the 50 % stress level.
· Table III (low resistance): failures DO occur at 50 % of yield within 30 days. 7075-T6 is in this table.
The same standard places 2024-T3 and T4 in Table III as well, while 6061 is in Table I in all tempers. One producer’s own rating points the same way: SCC rating C for T6/T651 and B for T73/T7351.

THE SHORT TRANSVERSE DIRECTION — the most critical detail. SCC resistance is direction-dependent, and in the words of the NASA standard “stress corrosion resistance of metals is worst in the short transverse direction“. The reason: in a rolled or extruded product the grains are elongated along the working direction; apply tension in the short transverse direction and the stress acts directly perpendicular to the long grain boundaries, so the crack can advance without cutting a single grain.
Practical consequences for the designer:
· When machining parts from heavy plate, avoid designs in which the TENSILE stress runs through the thickness (short transverse).
· Interference-fit bushings, over-torqued bolts and force-assembled parts produce exactly that kind of sustained tension — they are the classic starting points of SCC failures.
· Surfaces machined near the centre of a heavy plate cut the grain boundaries perpendicularly and are the most SCC-exposed surfaces of all.​‌​​‌​

THE SOLUTION IS T73 AND IT WORKS. Overageing coarsens and separates the η precipitates at the grain boundary and breaks the anodic path. The price is about 14 % of the proof stress (503 → 434 N/mm² typical). The gain is moving from Table III to Table I in the NASA classification — from a material that fails in 30 days at 50 % of yield to one that survives 30 days at 75 % of yield. And the fatigue strength does not change (158 N/mm²). Every 7075 part that sits outdoors, near the sea, in an interference fit or under sustained tension should be T73 or T7351.

WHERE IT FAILS — 2: exfoliation corrosion​‌​​‌​

This is a special form of intergranular corrosion seen in rolled 7075 products with elongated grains. Corrosion products form at the grain boundaries and occupy more volume than the metal did; that volume increase lifts the material like the pages of a book and splits it into leaves. From outside it appears as blistering and flaking; inside, the section loss has already happened.
T6 rolled product is susceptible. The T76 temper was developed for exactly this problem: it does not overage as far as T73, so it loses less strength, while still providing adequate resistance to exfoliation. The NASA standard places T76 and T7651 in Table I.

WHERE IT FAILS — 3: general corrosion, pitting and galvanic couples​‌​​‌​

Copper leaves copper-rich intermetallics on the surface and at the grain boundaries of 7075. Those particles are cathodic relative to the aluminium matrix and cause the surrounding matrix to dissolve preferentially. The result in a chloride environment is rapid pitting and intergranular corrosion along copper-depleted zones.
Galvanically, despite the copper, 7075 is still aluminium and is the ANODE against steel and stainless steel; but it also sits at a different potential from other aluminium alloys — putting 7075 in direct wet contact with 6082 or 5083 creates an aluminium-to-aluminium galvanic couple. See EN AW-5083 and EN AW-6082.

Protections: alclad, anodizing, primer​‌​​‌​

Frequently Asked Questions​‌​​‌​

Our specification says 7075-T6 but the part is outdoors and will carry an interference-fit bushing. What should we do?

Question the specification. This is the textbook scenario for an SCC failure.
All three conditions are present: (1) a susceptible microstructure — 7075-T6 is Table III, low resistance in NASA MSFC-STD-3029; (2) sustained tensile stress — that is the definition of an interference fit, and the stress runs hoop-wise around the bushing, which is often the short transverse direction; (3) a moist or chloride environment — outdoor service supplies that. The result of the trio is an intergranular crack that advances without warning.
The right answer is T73 or T7351. What you lose: typical proof stress from 503 to 434 N/mm², that is 13.7 %; tensile from 572 to 503. What you gain: the move from Table III to Table I in the same standard — from a material that fails in 30 days at 50 % of yield to one that survives 30 days at 75 % of yield. And the fatigue strength does not change (158 N/mm²), while elongation RISES (11 % → 13 %).
If the structural calculation cannot absorb a 14 % lower proof stress, three options remain: increase the section; move to 7050-T7451 (better particularly in heavy sections); or consider T76/T7651 — less strength loss than T73, less SCC resistance than T73.
And if T6 is insisted upon, these become mandatory: a clearance fit plus adhesive instead of an interference fit; shot peening in the bushing region; anodizing plus primer; scheduled inspection. But none of these substitutes for the change in microstructure.​‌​​‌​

Can 7075 be welded? Our supplier says “we do it with 5356”.

No. And “we do it with 5356” presents a crack-mitigation attempt as if it were a structural solution.
The ratings are unambiguous: one producer’s weldability rating is D (not recommended); on a European mill’s 1-6 scale gas, TIG and MIG fusion welding score 6 — the worst mark.
Three separate mechanisms:
1. Solidification (hot) cracking. According to a European mill sheet the freezing range of 7075 is 480-640 °C, a band 160 °C wide. In a pool solidifying across such a band, a liquid film remains at the grain boundaries at the last moment and shrinkage stress tears it. The Cu-Mg level of 7075 also sits at the peak of hot-cracking susceptibility.
2. HAZ liquation cracking. Beside the fusion line the parent metal’s grain boundaries partially melt. You cannot fix this by changing filler — the problem is the parent metal.
3. SCC in the weld region. Suppose you manage to lay the bead without cracking it: the weld metal and HAZ are effectively re-solutionised, cooled without control and naturally aged — that is, T6-like and open to SCC. Even a successful weld places the failure mode you were avoiding right beside the bead.
What is done instead: rivets and bolts (the standard aerospace answer; the rivet alloy is not 7075 but the 2017A/2024 class — see EN AW-2017A); structural adhesive bonding; and friction stir welding (FSW) — the only genuine “welding” option because nothing melts. Even in FSW the stir zone and its surroundings are re-solutionised and naturally aged; they are both weaker and stripped of SCC resistance.
If a welded high-strength aluminium structure is genuinely needed, the right address is the copper-free 7xxx alloys (7005 / 7020 / 7039) or 6082.​‌​​‌​

We want 150 mm thick 7075-T651 plate. Will that be a problem?

Yes — and this is the question that explains why 7050 exists.
The numbers: under EN 485-2 the minima for T651 plate are Rm ≥540 / Rp0.2 ≥460 at 6-12.5 mm, but they fall to Rm ≥360 / Rp0.2 ≥260 N/mm² at 120 mm and above. A 45 % loss in proof stress. So 150 mm 7075-T651 plate is not the material you expect from the thin plate table.
The cause is quench sensitivity. 7075 contains 0.18-0.28 % chromium and the Cr dispersoids act as heterogeneous nucleation sites during the quench. The core of a heavy plate cannot pass through the critical range (399-288 °C) fast enough (about 278 °C per second); the Zn and Mg that should have provided strength are spent as coarse, ineffective precipitates. Fracture toughness and SCC resistance fall the same way.
The right answer is 7050-T7451. 7050 uses zirconium instead of chromium; Zr dispersoids present far fewer nucleation sites and 7050 is markedly less quench sensitive. In the words of one producer source, 7050 “retains strength properties, fracture toughness levels and corrosion cracking resistance more effectively than other high-strength aluminium alloys, especially in thicker sections” and is widely used in aerospace in the 75-150 mm band (fuselage frames, bulkheads). 7050 also machines slightly better.
The residual-stress warning applies too: heavy 7075 plate carries high residual stress. Buying the stretched temper T651 is mandatory; in addition, remove material symmetrically and let the part rest between roughing and finishing.
Summary: in thin and medium sections 7075 is economical and correct; in heavy sections 7050 is the right answer.​‌​​‌​

There is a price difference between 7075-T6 and 7075-T73. Is it really necessary?

The answer depends on where and how the part will be stressed — and if you cannot answer that, buy T73.
T6 is adequate if: the part works indoors or in a dry, chloride-free environment; it is not under sustained tensile stress (cyclic load is not the driving force for SCC — SCC requires sustained stress); the section is thin and there is no tension in the short transverse direction; and it sits under a fully closed coating system.
T73/T7351 is MANDATORY if: the part is outdoors or near the sea; there is an interference fit, over-torque or forced assembly; it is machined from heavy plate and carries tension in the short transverse direction; it is a long-life part that is hard to inspect; or an aerospace/defence specification calls for SCC resistance.
Frame the price difference correctly. T73 costs more because it needs a two-stage and longer ageing cycle and because acceptance testing includes a conductivity measurement. But what you are buying is not strength — you are already giving up 14 % of it. What you are buying is the removal of a failure mode. Price the risk, not the alloy.
And an acceptance warning: when you buy T73, do not look only at hardness on the certificate. An under-overaged part can be close to T73 hardness but at T6 conductivity, and it is still SCC-susceptible. The T73 acceptance criterion is always the conductivity window (38-42 % IACS) AND the proof strength, checked together.​‌​​‌​

Common datasheet errors — check these before you order

1. MIXING THE T6 AND T73 VALUES — THE MOST COMMON AND MOST DANGEROUS ERROR. If a table presents the strength of T6 and the corrosion resistance of T73 in the same row, that table describes a material that does not exist. The reality: T6 typical 572/503, T73 typical 503/434 N/mm²; T6 is NASA Table III (low SCC resistance), T73 is Table I (high). You cannot have both.
2. “7075-T6 has excellent corrosion resistance.” WRONG. One producer’s general corrosion rating is C, a European mill’s seawater and weather rating is 4-5 (poor), and the NASA classification is Table III. 7075-T6 is not an unprotected outdoor material.
3. A narrow composition band. An encyclopaedia entry gives 5.6-6.1 % Zn, 2.1-2.5 % Mg, 1.2-1.6 % Cu; the specification limits are 5.1-6.1 % Zn, 2.1-2.9 % Mg, 1.2-2.0 % Cu. Do not reject a heat by mistaking the narrow band for the specification.
4. Melting temperature given as a single number. 532-635 °C and 480-640 °C appear for the same alloy in different sources. The low figure is the non-equilibrium eutectic melting point and it is the one that matters for solution treatment at 466 ± 6 °C. The wide freezing range is also the cause of the hot cracking.
5. Conflicting hardness values. For T73/T7351 one producer gives 130 HB and another 144 HB. Do not use hardness alone as the T73 acceptance criterion — use it with conductivity (38-42 % IACS).
6. Conflicting machinability rating. One producer summary gives D (poor) for 7075; a European mill gives 2 (good) for T651. The D rating most likely belongs to the ANNEALED (O) temper. Aged 7075 machines well; annealed 7075 does not.
7. Mixing the European minimum with the American typical in one table. EN 485-2 gives ≥540 / ≥460-470 (minimum) for T651 plate; American producer tables give 572 / 503 (typical) and 531 / 455 (minimum). Three sets of numbers, three meanings.
8. Heavy plate quoted with thin plate values. EN 485-2 gives 6-12.5 mm: 540/460 and ≥120 mm: 360/260. A 45 % loss in proof stress. That is the reason 7050 exists.
9. The claim “it can be welded with 5356”. The weldability rating is D / 6 (the worst). Hot cracking, HAZ liquation cracking and SCC in the weld region. A welded structural joint is not a solution for 7075.
10. W.Nr. confusion. 3.4365 = 7075 (AlZnMgCu1.5). Similar-looking numbers belong to other alloys; work from the EN AW number, not the old DIN name.
11. Mixing alclad values with bare sheet values. The cladding carries no load; the mechanical values of an alclad sheet are lower than those of bare sheet. And machining through the cladding ends the protection.
12. ASME / pressure-vessel claims. 7075 is not in ASME II Part D and will not be. There is no such thing as “ASME compliant 7075”.
13. Food-contact claims. 7075 is NOT suitable for food contact (1.2-2.0 % copper). A European mill sheet states this explicitly. Compare: 6082 is food-safe per DIN EN 602.
14. Writing T651 and T6 as a single row. Their mechanical values are identical but T651 is stretched. If you will machine precision parts from heavy plate, that is the difference between a flat part and a bowed one.
15. Assuming “overageing makes everything worse”. WRONG. In T73 elongation RISES (11 % → 13 %) and the fatigue strength DOES NOT CHANGE (158 N/mm²). What is lost is only proof and tensile strength.​‌​​‌​

​‌​​‌​