Skip to content
Materials and processes

You specced “304 stainless” because it says stainless. Eighteen months on a coastal railing, it’s pitting through.

Stainless isn’t one material, and steel isn’t one material either. Five grades cover most of what crosses a design review — S235JR, 304, 316, 6061 and 5052 — and picking between them by name recognition instead of by what actually differs is how a “corrosion-resistant” part corrodes.

What’s actually happening

“Stainless” describes a family, not a spec. 304 and 316 differ by one thing that matters most: 316 has 2–3% molybdenum, 304 doesn’t. That molybdenum is the entire reason 316 survives chlorides that pit 304 — everything else about the two alloys is close enough not to matter for most parts. Aluminium splits a similar way: 6061 is heat-treated to get its strength, which is also exactly why welding it locally undoes that strength; 5052 gets its strength from cold work alone, so it welds and bends without the same penalty.

Five grades, side by side

GradeTensile / yield (MPa)What it’s actually good atWhere it falls down
S235JR steel350–510 / 215–235Cheap, strong, welds easily, no alloy content to go wrongNo corrosion resistance at all — bare metal rusts
304 stainless~505 / ~215General-purpose corrosion resistance, indoor/mild environments, cost-effective stainlessPits and crevice-corrodes under sustained chlorides (coastal air, de-icing salt, washdown)
316 stainless~550 / ~240Chloride/marine resistance from the added molybdenumCosts noticeably more than 304 🔍 — no single reliable %-premium figure exists, market-dependent
6061 aluminium310 / 276Machinability, structural stiffness, heat-treated strengthLoses roughly 30–50% of that strength in the heat-affected zone when welded
5052 aluminium228 / 193Deep-drawn and bent sheet metal, welds without the HAZ penaltyLower as-supplied strength than 6061
Relative cost across all five: S235JR is cheapest, then 5052 and 6061 aluminium (5052 usually a little cheaper than 6061), then 304, then 316 most expensive. I could not find one consistent, current $/kg source covering all five at once 🔍 — commodity metal pricing moves too much and published figures disagreed on units and dates. Treat the ordering as directionally reliable, not a quote.

304 vs 316: the chloride problem

Both alloys rely on the same mechanism — a passive chromium-oxide layer a few nanometres thick that reforms itself when scratched, as long as oxygen can reach it. Chlorides attack that layer locally. 316’s molybdenum stabilises the passive layer against chloride attack; 304’s doesn’t, so 304 pits where 316 doesn’t. This is why 316 is the standard call for anything coastal, marine, or regularly washed down with salt — and why 304 is still the right, cheaper choice for most indoor and general enclosure work.

304 vs 316 under simulated chloride exposure
Pits visible — 3040
Pits visible — 3160
Illustrative relative behaviour, not measured corrosion-rate data.

6061 vs 5052: strength you can weld vs strength you can bend

6061 reaches its strength (T6 temper) through solution heat treatment and ageing. Weld it, and the heat near the joint locally reverses that ageing — the heat-affected zone can drop to roughly 60–70% of the parent metal’s strength. 5052 never relied on heat treatment for its strength in the first place — it’s strengthened by cold work (the H32 temper), so there’s no precipitation hardening to undo, and it comes out of a weld with far less of a strength penalty. The same logic runs the other way for forming: 6061’s greater hardness that makes it a better machining stock is what makes it a worse deep-draw candidate, and 5052 is the sheet-forming default for exactly that reason.

Strength across a welded joint — 6061 vs 5052
Parent metal strength
Strength at weld centreline
Retained
6061 HAZ range per published AA welding guidance (~50–70% of parent retained). 5052 figure from a TIG-welded 5052-H32 test coupon — illustrative, not a guaranteed value for every process.

Picking by environment

This is a teaching tool, not a spec generator — use it to sanity-check a first pick, not to replace a proper materials review.

Pick an environment — see which grade(s) fit

Why we care

Every one of these five grades is a correct choice somewhere and a mistake somewhere else. The failure mode isn’t picking a bad material — it’s picking a name you recognise instead of the one property that actually decides whether the part survives its environment or its manufacturing process. I ask two questions before specifying any of these: what’s it exposed to, and what’s going to be done to it before it’s a finished part. Those two answers usually make the grade obvious.

Related: Engineering Drawings (material + spec callout in the title block) · GD&T